Updated on 2022/11/28

写真a

 
OHYA, Jun
 
Scopus Paper Info  
Paper Count: 0  Citation Count: 0  h-index: 5

Citation count denotes the number of citations in papers published for a particular year.

Affiliation
Faculty of Science and Engineering, School of Creative Science and Engineering
Job title
Professor

Concurrent Post

  • Faculty of Science and Engineering   School of Fundamental Science and Engineering

  • Faculty of Science and Engineering   Graduate School of Creative Science and Engineering

  • Affiliated organization   Global Education Center

Research Institute

  • 2020
    -
    2022

    国際情報通信研究センター   兼任研究員

  • 2020
    -
    2022

    理工学術院総合研究所   兼任研究員

Education

  •  
    -
    1979

    University of Tokyo   Graduate School, Division of Engineering   Department of Precision Machinery Engineering  

  •  
    -
    1977

    University of Tokyo   Faculty of Engineering   Department of Precision Machinery Engineering  

Degree

  • University of Tokyo   Ph.D. (Dr. Engineering)-

Research Experience

  • 2014
    -
    Now

    Waseda Uniniversity, Department of Modern Mechanical Engineering   Professsor

  • 2002
    -
    2013

    present Waseda University, Graduate School of Global Information and Telecommunication Studies, Professor

  • 2005.07
    -
    2005.09

    Visiting professor, the University of Karlsruhe, Germany

  • 2000
    -
    2002

    Waseda University, Global Information and Telecommunication Institute, Professor

  • 1996
    -
    2000

    ATR Media Integration & Communication Research Laboratories, Department head

  • 1992
    -
    1996

    ATR Communication Systems Research Laboratories, Supervising researcher

  • 1989
    -
    1992

    NTT Human Interface Laboratories, Supervising researcher

  • 1988
    -
    1989

    University of Maryland, U.S.A.,Visiting research associate

  • 1979
    -
    1988

    NTT Electrical Communications Laboratories,Researcher

▼display all

Professional Memberships

  •  
     
     

    The Institute of Image Electronics Engineers of Japan

  •  
     
     

    The Virtual Reality Society of Japan

  •  
     
     

    Information Processing Society of Japan

  •  
     
     

    The Institute of Electronics, Information and Communication Engineers

  •  
     
     

    Institute of Electronics, Information and Communication Engineers

 

Research Areas

  • Intelligent informatics

Research Interests

  • Computer vision,Computer graphics,Virtual reality,Computer Vision,Virtual Reality,Multimedia Tele-communication

Papers

  • GEAR: A Graph Edge Attention Routing Algorithm Solving Combinatorial Optimization Problem with Graph Edge Cost

    Yuhei Senuma, Zhao Wang, Yuusuke Nakano, Jun Ohya

    Proceedings of the 10th ACM SIGSPATIAL International Workshop on Analytics for Big Geospatial Data (BigSpatial '22)     8 - 16  2022.11  [Refereed]

  • Temperature-controlled laser thermal therapy system using a newly developed laparoscopic system equipped with an ultra-compact thermographic camera

    Manabu Harada, Yuji Morimoto, Ohara Mutsuki, Jun Ohya, Ken Masamune, Yujiro Itazaki, Takao Sugihara, Hironori Tsujimoto, Yoji Kishi, Hideki Ueno

    Scientific Reports   12 ( 1 )  2022.10  [Refereed]

     View Summary

    Abstract

    Laser thermal therapy is one of the treatments for malignant tumors. We developed a thermal endoscope using an ultra-compact thermo-sensor and established a new laparoscopic laser thermal therapy system to heat cancer tissue at an appropriate temperature, focusing on the fact that thermographic cameras are capable of two-dimensional temperature mapping. Hepatocellular carcinoma (N1S1) cells were implanted into the livers of Sprague–Dawley rats (n = 13) to create orthotopic hepatocellular carcinoma. Six of the rats underwent laparoscopic laser thermotherapy (70 °C, 5 min) using the newly developed system, and the others underwent laparoscopic insertion only. Lesion volume measurement and histological evaluation were performed in all of the rats. The laparoscopic laser thermal therapy system provided stable temperature control. When a temperature of 70 °C was used for the set temperature, the temperature of the target cancer was maintained within the range of 68–72 °C for 93.2% of the irradiation time (5 min). The median volume of the tumors that were thermally treated was significantly smaller than that of the untreated tumors. The newly developed laparoscopic laser thermal therapy system was capable of maintaining the temperature of the tumor surface at any desired temperature and was proven to be effective in treatment of the rat hepatocellular carcinoma model.

    DOI

    Scopus

  • Preliminary Investigation of Collision Risk Assessment with Vision for Selecting Targets Paid Attention to by Mobile Robot

    Masaaki Hayashi, Tamon Miyake, Mitsuhiro Kamezaki, Junji Yamato, Kyosuke Saito, Taro Hamada, Eriko Sakurai, Shigeki Sugano, Jun Ohya

    Proceedings of 2022 31st IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)     624 - 629  2022.08  [Refereed]

  • Automatic Diagnosis of Early-Stage Oral Cancer and Precancerous Lesions from ALA-PDD Images Using GAN and CNN

    Taro Fujimoto, Eiji Fukuzawa, Seiko Tatehara, Kazuhito Satomura, Jun Ohya

    2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)     2161 - 2164  2022.07  [Refereed]

  • Context Enhanced Traffic Segmentation: traffic jam and road surface segmentation from aerial image

    Yubo Wang, Zhao Wang, Yuusuke Nakano, Ken Nishimatsu, Katsuya Hasegawa, Jun Ohya

    2022 IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)     1 - 5  2022.06  [Refereed]

  • Efficient landslide detection by UAV-based multi-temporal visual analysis

    Yosuke Yamaguchi, Kai Matsui, Jun Ohya, Katsuya Hasegawa, Hiroshi Nagahashi

    Proc. IS&T Int’l. Symp. on Electronic Imaging: Intelligent Robotics and Industrial Applications using Computer Vision   34   307-1 - 307-6  2022.01  [Refereed]

  • Estimating the 3D Cut Position of Pork Frontal Legs in RGBD Images by a Deep Learning Based Method for Achieving a Robot That Cuts Pork Legs Autonomously

    Ryo Onuki, Taiki Suzuki, Jeonghwang Hayashi, Chanjin Seo, Jun Ohya, Takaaki Ohkawauchi

    The 7th IIEEJ International Conference on Image Electronics and Visual Computing   ( 3A-4 ) 1 - 4  2021.09  [Refereed]

  • Early Detection of Objects on the Road Using V2V and Multiple Image Streams for Supervised Deep-Learning Based Autonomous Driving

    Zelin Zhang, Jun Ohya

    The 7th IIEEJ International Conference on Image Electronics and Visual Computing   ( 3A-3 ) 1 - 4  2021.09  [Refereed]

  • Interpreting Evaluation Criteria of Running Foot Strike Classifier Using Kinematic Model

    Chanjin Seo, Yuta Goto, Koji Tagami, Hiroyuki Ogata, Kazuyuki Kanosue, Jun Ohya

    The 7th IIEEJ International Conference on Image Electronics and Visual Computing   ( 1A-4 ) 1 - 4  2021.09  [Refereed]

  • Movement Control with Vehicle-to-Vehicle Communication by Using End-to-End Deep Learning for Cooperative Autonomous Driving

    Zelin, Zhang, Jun OHYA

    10th International Conference on Pattern Recognition Applications and Methods     377 - 385  2021.02  [Refereed]

  • Quantitative Method for Evaluating the Coordination between Sprinting Motions using Joint Coordinates Obtained from the Videos and Cross-correlations

    Masato Sabanai, Chanjin Seo, Hiroyuki Ogata, Jun Ohya

    10th International Conference on Pattern Recognition Applications and Methods     531 - 539  2021.02  [Refereed]

     View Summary

    This paper proposes a method for quantitatively evaluating sprinting motions using the videos of runners. Specifically, this paper explores the coordination between physical motions, which has been recognized as very important in sprinting. After detecting and normalizing the joint coordinates from sprinting videos, the cross-correlations of two windowed time-series data are calculated using the windowing cross-correlation function, and the coordination between the motions of the two joints is quantified. Experiments that use 20 subjects are conducted. As a result of classifying the cross-correlation obtained from the subjects' data into two clusters using k-means clustering, conditions in which the obtained cluster includes a high percentage of inexperienced sprinters are found. To verify whether the motions corresponding to these conditions are valid as the evaluation criterion of sprinting, Spearman's rank correlation coefficients between cross-correlations and 30-m time records are calculated. The results show a weak correlation with respect to the coordination between the elbow and knee motions. Therefore, it can be said that the cross-correlation corresponding to the coordination can be used as a quantitative criterion in sprinting.

  • Study of a Method for Detecting Dominant Vegetation in a Field from RGB Images Using Deep Learning in Synecoculture Environment

    SOYA Kanta, AOTAKE Shuntaro, OGATA Hiroyuki, OHYA Jun, OHTANI Takuya, TAKANISHI Atsuo, FUNABASHI Masatoshi

    Proceedings of the Annual Conference of the Institute of Image Electronics Engineers of Japan   49   57 - 57  2021

     View Summary

    Synecoculture ™ is a method of farming that produces useful plants while making multifaceted use of the
    self organizing ability of the ecosystem by growing a wide variety of plants densely mixed in the same
    farmland. As a technology to support Synecoculture , robotics are being developed to automate major
    management tasks Still, the complexity of recognition and operation is imposing a heavy burden against
    automation compared with conventional farming that is based on a uniform operation of a single plant.
    On Synecoculture it is essential to grow plants with high diversity , but the dominance of some plants over
    other s may change the species composition and occupancy in the ecosystem which might result in reduce d
    diversity Pruning these excessively dominant plants is needed to maintain the balance of species
    composition in the vegetation of Synecoculture . In this study, we aim to detect such overly propagating plants
    that m ight reduce the diversity of the vegetation community (dominant plants).
    The proposed method detects the dominant plants using the Chopped Picture Method (CPM), a
    Convolutional Neural Network CNN learning method for segmenting RGB images. In this study, we treat
    Mentha suaveolens ( as one of the dominant plants to be detected and trained the CNN with three
    labels: “mint,” “plants other than mint” and “others.” As a result, we obtained high accuracy segmentation
    in detecting the dominant plants, especially in distinguishing the plant group from the non plant group.

    DOI

  • Study of a Method for Recognizing Field Covering Situation by Applying Semantic Segmentation to RGB Images in Synecoculture Environment

    YOSHIZAKI Reina, AOTAKE Shuntaro, OGATA Hiroyuki, OHYA Jun, OHTANI Takuya, TAKANISHI Atsuo, FUNABASHI Masatoshi

    Proceedings of the Annual Conference of the Institute of Image Electronics Engineers of Japan   49   59 - 59  2021

     View Summary

    Synecoculture™ is a method of farming that produces useful plants while making multifaceted use of the self-organizing ability of the ecosystem by growing a wide variety of plants densely mixed in the same farmland. As a technology to support Synecoculture, robotics are being developed to automate major management tasks. Still, the complexity of recognition and operation is imposing a heavy burden against automation compared with conventional farming that is based on a uniform operation of a single plant.
    In Synecoculture, it is essential to cover the topsoil with vegetation. If the topsoil is exposed, it is necessary to introduce seeds and seedlings to fill the gap with vegetation. In this study, we aim to recognize the area of the bare soil surface with pixel-wise precision.
    In the proposed method, each pixel segments into two classes: “vegetation” or “no vegetation.” by applying semantic segmentation to RGB images with the Focal Loss function. By comparing accuracy with different values of parameters for the semantic segmentation, our approach showed that this method could achieve high accuracy with a relatively small number of images for training.

    DOI

  • Data Collection Through Translation Network Based on End-to-End Deep Learning for Autonomous Driving

    Zelin, Zhang, Jun OHYA

    IS&T International Symposium on Electronic Imaging. (EI 2021)     115 - 121  2021.01  [Refereed]

  • Extracting and Interpreting Unknown Factors with Classifier for Foot Strike Types in Running

    Chanjin Seo, Masato Sabanai, Yuta Goto, Koji Tagami, Hiroyuki Ogata, Kazuyuki Kanosue, Jun Ohya

    25th International Conference on Pattern Recognition (ICPR)     3217 - 3224  2021.01  [Refereed]

     View Summary

    This paper proposes a method that can classify foot strike types using a deep learning model and can extract unknown factors, which enables to evaluate running motions without being influenced by biases of sports experts, using the contribution degree of input values (CDIV). Accelerometers are attached to the runner's body, and when the runner runs, a fixed camera observes the runner and acquires a video sequence synchronously with the accelerometers. To train a deep learning model for classifying foot strikes, we annotate foot strike acceleration data for RFS (Rearfoot strike) or non-RFS objectively by watching the video. To interpret the unknown factors extracted from the learned model, we calculate two CDIVs: the contributions of the resampling time and the accelerometer value to the output (foot strike type). Experiments on classifying unknown runners' foot strikes were conducted. As a common result to sport science, it is confirmed that the CDIVs contribute highly at the time of the right foot strike, and the sensor values corresponding to the right and left tibias contribute highly to classifying the foot strikes. Experimental results show the right tibia is important for classifying foot strikes. This is because many of the training data represent difference between the two foot strikes in the right tibia. As a conclusion, our proposed method could extract unknown factors from the classifier and could interpret the factors that contain similar knowledge to the prior knowledge of experts, as well as new findings that are not included in conventional knowledge.

    DOI

    Scopus

  • Estimating the Orientation of 3D-Structured Switches Using the RGB-D Camera Attached to a Disaster Response Robot

    Takuya Kanda, Li Qi, Jun Ohya, Atsuo Takanishi, Hiroyuki Ogata, Kenji Hshimoto

    International Journal of Mechanical Engineering and Robotics Research   9 ( 10 ) 1432 - 1438  2020.10  [Refereed]

  • Developing Thermal Endoscope for Endoscopic Photothermal Therapy for Peritoneal Dissemination

    Mutsuki Ohara, Sohta Sanpei, Chanjin Seo, Jun Ohya, Ken Masamune, Hiroshi Nagahashi, Yuji Morimoto, Manabu Harada

    2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)     3040 - 3047  2020.10  [Refereed]

     View Summary

    As a novel therapy for peritoneal dissemination, it is desired to actualize an endoscopic photothermal therapy, which is minimally invasive and is highly therapeutically effective. However, since the endoscopic tumor temperature control has not been actualized, conventional therapies could damage healthy tissues by overhearing. In this paper, we develop a thermal endoscope system that controls the tumor temperature so that the heated tumor gets necrotic. In fact, our thermal endoscope contains a thermal image sensor, a visible light endoscope and a laser fiber. Concerning the thermal image sensor, the conventional thermal endoscope has the problem that the diameter is too large, because the conventional endoscope loads a large thermal image sensor with high-resolution. Therefore, this paper uses a small thermal image sensor with low resolution, because the diameter of the thermal endoscope needs to be smaller than 15mm in order to be inserted into the trocar. However, this thermal image sensor is contaminated by much noise. Thus, we develop a tumor temperature control system using a feedback control and tumor temperature estimation based on Gaussian function, so that the noisy, small thermal image sensor can be used. As experimental results of the proposed endoscopic photothermal therapy for the hepatophyma carcinoma model of rats, it turns out that the tumor temperature by which the heated tumor gets necrotic can be kept stable. It can be said that our endoscopic photothermal therapy achieves a certain degree of therapy effect.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Automatic Estimation of the Position and Orientation of Stairs to Be Reached and Climbed by a Disaster Response Robot by Analyzing 2D Image and 3D Point Cloud

    Kazuya Miyakawa, Takuya Kanda, Jun Ohya, Hiroyuki Ogata, Kenji Hashimoto, Atsuo Takanishi

    International Journal of Mechanical Engineering and Robotics Research   9 ( 9 ) 1312 - 1321  2020.09  [Refereed]

  • Classification of Aortic Stenosis Using ECG by Deep Learning and its Analysis Using Grad-CAM

    Erika Hata, Chanjin Seo, Masafumi Nakayama, Kiyotaka Iwasaki, Takaaki Ohkawauchi, Jun Ohya

    2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)     1548 - 1551  2020.07  [Refereed]

    DOI

  • Locating Mechanical Switches Using RGB-D Sensor Mounted on a Disaster Response Robot

    Takuya Kanda, Kazuya Miyakawa, Jeonghwang Hayashi, Jun Ohya, Hiroyuki Ogata, Kenji Hashimoto, Xiao Sun, Takashi Matsuzawa, Hiroshi Naito, Atsuo Takanishi

    Electronic Imaging 2020, Intelligent Robotics and Industrial Applications using Computer Vision 2020   IRIACV ( 16 ) 1 - 6  2020.01  [Refereed]

  • Makeup Interpolation Based on Color and Shape Parametrization

    Jieun Cho, Jun Ohya, Sang Il Park

    Symmetry   11 ( 1422 ) 1 - 15  2019.11  [Refereed]

  • Disaster Response Robot’s Autonomous Manipulation of Valves in Disaster Sites Based on Visual Analyses of RGBD Images

    Keishi Nishikawa, Asaki Imai, Kazuya Miyakawa, Takuya Kanda, Takashi Matsuzawa, Kenji Hashimoto, Atsuo Takanishi, Hiroyuki Ogata, Jun Ohya

    2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)     4790 - 4797  2019.11  [Refereed]

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Investigating Relationship between Running Motions and Skills Acquired from Jump Trainings,

    Chanjin Seo, Masato Sabanai, Hiroyuki Ogata, Jun Ohya

    icSPORTS     198 - 203  2019.09  [Refereed]

     View Summary

    To identify the difference in performers' motions, this paper investigates the relationship between running motions and the result of evaluating motions during jump training. To clarify the relationship, two experiments were performed using 17 subjects as follows: i) obtaining sequences of human joints during running to evaluate running motions, and ii) obtaining motions during jump training which could skill up the running motions. According to the result of those experiments, we confirmed that whether a running motion is good or not relies greatly on the number of acquired skills.

    DOI

    Scopus

  • A Criterion that Evaluates the Sprinting Motion Using the Joint Coordinates Detected from the Video and Machine Learning

    Masato Sabanai, Chanjin Seo, Hiroyuki Ogata, Jun Ohya

    The 6th IIEEJ International Conference on Image Electronics and Visual Computing (IEVC 2019)   ( 3C-3 ) 1 - 4  2019.08  [Refereed]

  • Local Branch Ensemble Network: Autonomous Driving System Using End-to-End Deep Learning

    Zelin Zhang, Jun Ohya

    The 6th IIEEJ International Conference on Image Electronics and Visual Computing (IEVC 2019)   ( 2C-3 ) 1 - 4  2019.08  [Refereed]

  • Applying Curvatures Estimated from 3D Point Clouds to Environment Recognition in Forests Using SegNet

    Takeo Kaneko, Hiroyuki Fujimoto, Junya Morimoto, Takuya Hayashi, Junji Yamato, Jun Ohya

    The 6th IIEEJ International Conference on Image Electronics and Visual Computing (IEVC 2019)   ( 2C-4 ) 1 - 4  2019.08  [Refereed]

  • A Method of Proposing the Tracking Suitable for Supplementing an Exerciser’s Lacking Skill in Sprint

    Chanjin Seo, Masato Sabanai, Jun Ohya, Hiroyuki Ogata

    The 6th IIEEJ International Conference on Image Electronics and Visual Computing (IEVC 2019)   ( 3C-2 ) 1 - 4  2019.08  [Refereed]

  • Region Mining of Fetal Head in Ultrasound Image Based on Weakly Supervised Annotations and Deep Learning

    Yan LI, Rong XU, Artus KROHN-GRIMBERGHE, Jun OHYA, Hiroyasu IWATA

    IIEEJ Transactions on Image Electronics and Visual Computing   7 ( 1 ) 46 - 51  2019.06  [Refereed]

  • Deep Learning Based Uterus Localization and Anatomical Structure Segmentation on Fetal Ultrasound Image

    Yan LI, Rong XU, Artus KROHN-GRIMBERGHE, Jun OHYA, Hiroyasu IWATA

    IIEEJ Transactions on Image Electronics and Visual Computing   7 ( 1 ) 13 - 23  2019.06  [Refereed]

  • An Algorithm for Tracking Guitarists’ Fingertips Based on CNN-Segmentation and ROI Associated Particle Filter

    Zhao WANG, Jun OHYA

    Journal of Imaging Science and Technology   63 ( 2 ) 20506-1 - 20506-9  2019.03  [Refereed]

  • Detecting a Fetus in Ultrasound Images using Grad CAM and Locating the Fetus in the Uterus

    Genta Ishikawa, Rong Xu, Jun Ohya, Hiroyasu Iwata

    International Conference on Pattern Recognition Application and Methods     181 - 189  2019.02  [Refereed]

  • Understanding Sprinting Motion Skills Using Unsupervised Learning for Stepwise Skill Improvements of Running Motion

    Chanjin Seo, Masato Sabanai, Hiroyuki Ogata, Jun Ohya

    International Conference on Pattern Recognition Application and Methods     467 - 475  2019.02  [Refereed]

     View Summary

    To improve running performances, each runner's skill, such as characteristics and habits, needs to be known, and feedback on the performance should be outputted according to the runner's skill level. In this paper, we propose a new coaching system for detecting the skill of a runner and a method of giving feedback using a sprint motion dataset. Our proposed method calculates an extracted feature to detect the skill using an autoencoder whose middle layer is an LSTM layer; we analyse the feature using hierarchical clustering, and we analyse the human joints that affect the skill. As a result of experiments, five clusters are obtained using hierarchical clustering. This paper clarifies how to detect the skill and to output feedback to achieve a level of performance one step higher than the current level.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Detecting and Tracking Surgical Tools for Recognizing Phases of the Awake Brain Tumor Removal Surgery.

    Hiroki Fujie, Keiju Hirata, Takahiro Horigome, Hiroshi Nagahashi, Jun Ohya, Manabu Tamura, Ken Masamune, Yoshihiro Muragaki

    International Conference on Pattern Recognition Application and Methods     190 - 199  2019  [Refereed]

    DOI

    Scopus

  • Automatic Estimation of the Position and Orientation of the Drill to Be Grasped and Manipulated by the Disaster Response Robot Based on Analyzing Depth Camera Information

    Keishi Nishikawa, Jun Ohya, Hiroyuki Ogata, Kenji Hashimoto, Takashi Matsuzawa, Asaki Imai, Shunsuke Kimura, Atsuo Takanishi

    IS&T International Symposium on Electronic Imaging 2019, Intelligent Robotics and Industrial Applications using Computer Vision 2019   IRIACV ( 452 ) 1 - 6  2019.01  [Refereed]

  • Pattern and Frontier-based, Efficient and Effective Exploration of Autonomous Mobile Robots in Unknown Environments

    Hiroyuki FUJIMOTO, Junya MORIMOTO, Takuya HAYASHI, Junji YAMATO, Hiroyuki ISHII, Jun OHYA, Atsuo TAKANISHI

    IS&T International Symposium on Electronic Imaging 2019, Autonomous Vehicles and Machines Conference 2019     039-1 - 039-6  2019.01  [Refereed]

  • An Accurate and Robust Method for Detecting Fetal Heads in Ultrasound Images Based on Integrating a Voting Scheme and an Improved IRHT

    Genta Ishikawa, Rong Xu, Jun Ohya, Hiroyasu Iwata

    IIEEJ Transactions on Electronics and Visual Computing,   6 ( 2 ) 65 - 73  2018.12  [Refereed]

  • Automatic detection of valves with disaster response robot on basis of depth camera information

    Keishi Nishikawa, Jun Ohya, Hiroyuki Ogata, Kenji Hashimoto, Takashi Matsuzawa, Atsuo Takanishi

    Digital Image Computing: Techniques and Applications (DICTA 2018),     253 - 258  2018.12  [Refereed]

  • A 3D guitar fingering assessing system based on CNN-Hand pose estimation and SVR-Assessment

    Zhao Wang, Jun Ohya

    IS and T International Symposium on Electronic Imaging Science and Technology   Part F138660   2781 - 2785  2018.01  [Refereed]

     View Summary

    © 2018, Society for Imaging Science and Technology. This paper proposes a guitar fingering assessing system based on CNN (Convolutional Neural Network) hand pose estimation and SVR (Support Vector Regression) evaluation. To spur our progress, first, a CNN architecture is proposed to estimate temporal 3D position of 16 joints of hand; then, based on a DCT (Discrete Cosine Transform) feature and SVR, fingering of guitarist is scored to interpret how well guitarist played. We also release a new dataset for professional guitar playing analysis with significant advantage in total number of video, professional judgement by expert of guitarist, accurate annotation for hand pose and score of guitar performance. Experiments using videos containing multiple persons' guitar plays under different conditions demonstrate that the proposed method outperforms the current state-of-art with (1) low mean error (Euclid distance of 6,1 mm) and high computation efficiency for hand pose estimation; (2) high rank correlation (0.68) for assessing the fingering (C major scale and symmetrical excise) of guitarists.

    DOI

    Scopus

  • Estimating the UAVs' crash point based on optical flows' voting in the image plane

    Yusaku Hatano, Hiroshi Nagahashi, Chen Yi, Jun Ohya

    IS and T International Symposium on Electronic Imaging Science and Technology   Part F138654  2018.01  [Refereed]

     View Summary

    © 2018, Society for Imaging Science and Technology. Towards the actualization of an air bag system for the UAV's crash, this paper proposes a method for estimating the UAV's crash site from the video sequence acquired by the camera attached to the UAV. The crash point can be considered to correspond to the divergence point of the optical flows. In the accumulator, the cells at which the optical flows (straight lines) pass through are incremented by one. After performing this process for all the optical flows, the cell with the largest vote is obtained as the crash point (divergence point) in the image plane. Experiments using a hand held camera show that the accuracy of estimating the crash site is increased as the camera approaches the target plane. Overall, the experimental results are promising.

    DOI

    Scopus

  • An accurate and robust algorithm for tracking guitar neck in 3D based on modified RANSAC homography

    Zhao Wang, Jun Ohya

    IS and T International Symposium on Electronic Imaging Science and Technology   Part F138651  2018.01  [Refereed]

     View Summary

    © 2018 Society for Imaging Science and Technology. Towards the actualization of an automatic guitar teaching system that can supervise guitar players, this paper proposes an algorithm for accurately and robustly tracking the 3D position of the fretboard from the video of guitar plays. First, we detect the SIFT features within the guitar fretboard and then match the detected points using KD-tree searching based matching algorithm frame by frame to track the whole fretboard. However, during the guitar plays, due to movements of the guitar neck or occlusions caused by guitar players' fingers, the feature points on the fretboard cannot always be matched accurately even though applying traditional RANSAC homography. Therefore, by using our modified RANSAC algorithm to filter out the matching error of the feature points, perspective transformation matrix is obtained between the correctly matched feature points detected at the first and other frames. Consequently, the guitar neck is tracked correctly based on the perspective transformation matrix. Experiments show promising results such as high accuracy: the total mean tracking error of only 4.17 mm and variance of 1.5 for the four tracked corners of the fretboard. This indicates the proposed method outperforms related tracking works including state-of-art Fully-convolutional Network.

    DOI

    Scopus

  • Discriminating the presence of the cerebral aneurysm using shape features obtained from medical images of the cerebral vessel

    Kosei Kikuchi, Takanobu Yagi, Xu Rong, Jun Ohya

    IS and T International Symposium on Electronic Imaging Science and Technology   Part F138660   2831 - 2836  2018.01  [Refereed]

     View Summary

    © 2018, Society for Imaging Science and Technology. Towards the establishment of the preventive medical care for the cerebral aneurysm, this paper proposes an SVM based method for building a discrimination function that classifies the presence or absence of the cerebral aneurysm using the cerebral blood vessel's shape features obtained from medical images such as MR images. Using the discrimination function, this paper explores how much each feature affects the onset of the cerebral aneurysm. This paper deals with the internal carotid artery (ICA). The blood vessel (ICA)'s shape features are extracted from medical images of 18 persons without cerebral aneurysm and 13 patients with a cerebral aneurysm. From the medical image, the cross sections and centerline of the ICA are obtained. The cross sections are divided into nine sections along the centerline. Shape features such as the cross sectional area, its circularity, curvature, torsion, length of the centerline and branch angles are obtained in each section; as a total, 113 features including the mean and variance of some features in each section are used for building the SVM. As a result of conducting the experiments, the accuracy for discriminating the presence/absence of the aneurysm by the SVM is 90.3%. In the obtained discrimination function, the coefficient values of the function can be considered how much the features affect the onset of the aneurysm. The features that could significantly cause the onset of the cerebral aneurysm are clarified, and the reasons why these features are significant are discussed.

    DOI

    Scopus

  • An image processing based method for chewing detection using Variable-intensity template

    Atsuto Fujimoto, Takaaki Ohkawauchi, Junji Yamato, Jun Ohya

    IS and T International Symposium on Electronic Imaging Science and Technology   Part F138660   2791 - 2794  2018.01  [Refereed]

     View Summary

    © 2018, Society for Imaging Science and Technology. To contribute to the actualization of the care worker assistance robot, this paper proposes a method for detecting whether the care receiver is chewing from the video sequence acquired by the camera that observes that receiver. The proposed method detects the receiver's face and areas for both cheeks and chin. After applying some normalization to the areas, chewing detection that uses a variable-intensity template is performed, where the template consists of shape models, interest points and intensity distribution model. A likelihood based on the variableintensity template is computed so that the receiver is judged whether the receiver is chewing. Experiments using seven subjects are conducted. As a result, the accuracy of chewing detection by the proposed method is 83%, which is quite promising.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Exploring the effectiveness of using temporal order information for the early-recognition of suture surgery's six steps based on video image analyses of surgeons' hand actions

    Miwa Tsubota, Ye Li, Jun Ohya

    RO-MAN 2017 - 26th IEEE International Symposium on Robot and Human Interactive Communication   2017-   464 - 469  2017.12  [Refereed]

     View Summary

    To alleviate the recent shortage problem of nurses, the actualization of RSN (Robotic Scrub Nurse) that can autonomously judge the current step of the surgery and pass the surgical instruments needed for the next step to surgeons is desired. The authors developed a computer vision based algorithm that can early-recognize only two steps of suture surgery. Based on the past work, this paper explores the effectiveness of utilizing temporal order of the six steps in suture surgery for the early-recognition. Our early-recognition algorithm consists of two modules: start point detection and hand action early-recognition. Segments of the test video that start from each quasi-start point are compared with the training data, and their probabilities are calculated. According to the calculated probabilities, hand actions could be early-recognized. To improve the early-recognition accuracy, temporal order information could be useful. This paper checks confusions of three steps' early recognition results, and if necessary, early-recognizes again after eliminating the wrong result, while for the other three steps, temporal order information is not utilized. Experimental results show our early-recognition method that utilizes the temporal order information achieves better performances.

    DOI

    Scopus

  • Automatic fetal body and amniotic fluid segmentation from fetal ultrasound images by encoder-decoder network with inner layers

    Yan Li, Rong Xu, Jun Ohya, Hiroyasu Iwata

    Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS     1485 - 1488  2017.09  [Refereed]

     View Summary

    This paper explores the effectiveness of applying a deep learning based method to segment the amniotic fluid and fetal tissues in fetal ultrasound (US) images. The deeply learned model firstly encodes the input image into down scaled feature maps by convolution and pooling structures, then up-scale the feature maps to confidence maps by corresponded un-pooling and convolution layers. Additional convolution layers with 1×1 sized kernels are adopted to enhance the feature representations, which could be used to further improve the discriminative learning of our model. We effectively update the weights of the network by fine-tuning on part of the layers from a pre-trained model. By conducting experiments using clinical data, the feasibility of our proposed approach is compared and discussed. The result proves that this work achieves satisfied results for segmentation of specific anatomical structures from US images.

    DOI PubMed

    Scopus

    21
    Citation
    (Scopus)
  • Automatic, Accurate Estimation of the Position and Pose of a Ladder in 3D Point Cloud

    Keishi Nishikawa, Zhao Wang, Jun Ohya, Takashi Matsuzawa, Kenji Hashimoto, Atsuo Takanishi

    The IIEEJ Image Electronics and Visual Computing Workshop 2017   ( 5C-2 ) 1 - 6  2017.03  [Refereed]

  • An Accurate and Robust Fetal Head Detection Algorithm Integrating a Voting Scheme and an Improved IRHT Method

    Genta Ishikawa, Rong Xu, Jun Ohya, Hiroyasu Iwata

    The IIEEJ Image Electronics and Visual Computing Workshop 2017   ( 5C-3 ) 1 - 5  2017.03  [Refereed]

  • Pregnant Uterine Ultrasound Image Segmentation by Encoding-Decoding Convolutional Neural Network

    Yan Li, Rong Xu, Jun Ohya, Hiroyasu Iwata

    The IIEEJ Image Electronics and Visual Computing Workshop 2017   ( 1C-1 ) 1 - 4  2017.03  [Refereed]

  • Fingertips tracking algorithm for guitarist based on temporal grouping and pattern analysis

    Zhao Wang, Jun Ohya

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   10118   212 - 226  2017  [Refereed]

     View Summary

    This paper proposes a temporal Grouping and pattern analysis-based algorithm that could track the fingertips of guitarists during their guitar playing towards the actualization of the automatic guitar fingering recognition system. First a machine learning-based Bayesian Pixel Classifier is used to segment the hand area on the test data. Then, the probability map of fingertip is generated on the segmentation results by counting the voting numbers of the Template Matching and Reversed Hough Transform. Furthermore, a temporal Grouping algorithm, which is a geometry analysis for consecutive frames, is applied to removal noise and group the same fingertips (index finger, middle finger, ring finger, little finger). Then, a data association algorithm is utilized to associate 4 tracked fingers (index finger, middle finger, ring finger, little finger) with their correspondent tracked results frame by frame. Finally, particles are distributed only between the associated fingertip candidates to track the fingertips of guitarist effectively. The experimental result demonstrates that this fingertip tracking algorithm is robust enough for tracking fingertips (1) without any constrains such us color marker
    (2) under the complex contexts, such us complicated background, different illumination conditions, (3) with the high tracking accuracy (mean error 3.36 pixels for four fingertips).

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Fingertips tracking algorithm for guitarist based on temporal grouping and pattern analysis

    Zhao Wang, Jun Ohya

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   10118   212 - 226  2017  [Refereed]

     View Summary

    This paper proposes a temporal Grouping and pattern analysis-based algorithm that could track the fingertips of guitarists during their guitar playing towards the actualization of the automatic guitar fingering recognition system. First a machine learning-based Bayesian Pixel Classifier is used to segment the hand area on the test data. Then, the probability map of fingertip is generated on the segmentation results by counting the voting numbers of the Template Matching and Reversed Hough Transform. Furthermore, a temporal Grouping algorithm, which is a geometry analysis for consecutive frames, is applied to removal noise and group the same fingertips (index finger, middle finger, ring finger, little finger). Then, a data association algorithm is utilized to associate 4 tracked fingers (index finger, middle finger, ring finger, little finger) with their correspondent tracked results frame by frame. Finally, particles are distributed only between the associated fingertip candidates to track the fingertips of guitarist effectively. The experimental result demonstrates that this fingertip tracking algorithm is robust enough for tracking fingertips (1) without any constrains such us color marker
    (2) under the complex contexts, such us complicated background, different illumination conditions, (3) with the high tracking accuracy (mean error 3.36 pixels for four fingertips).

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Subaction Based Early Recognition of Surgeons’ Hand Actions from Continuous Surgery Videos

    Ye Li, Jun Ohya, Toshio Chiba, Xu Rong, Hiromasa Yamashita

    IIEEJ Transactions on Image Electronics and Visual Computing   4 ( 2 ) 124 - 135  2016.12  [Refereed]

    CiNii

  • Supervoxel-Based Detection and Segmentation of Moving Objects from Dynamic RGBD Images

    Naotomo Tatematsu, Jun Ohya, Larry S. Davis

    IIEEJ Transactions on Image Electronics and Visual Computing   4 ( 1 ) 32 - 42  2016.06  [Refereed]

    CiNii

  • Computer Vision Technologies Towards the Actualization of Robotic Scrub Nurses

    LI Ye, OHYA Jun

    NIHON GAZO GAKKAISHI (Journal of the Imaging Society of Japan)   55 ( 1 ) 106 - 110  2016

     View Summary

    Robotic Scrub Nurse (RSN) is a developing research and a challenging issue in the field of medical robotics. Audio processing, wearing-sensor and computer vision are applied to control the robot that could support surgeons during surgeries as an alternative of human nurse. On the other hand, surgeons wish that RSN should be able to decide the correct action by recognizing surgical situations autonomously just like skilled human nurses. In general, surgical situations can be discriminated by surgeons hand actions during surgeries. This article introduces and explains the authors computer vision based method for recognizing surgical hand actions of surgeons from video sequences.

    DOI CiNii

  • Tracking the guitarist’s fingers as well as recognizing pressed chords from a video sequence

    Zhao Wang, Jun Ohya

    IS and T International Symposium on Electronic Imaging Science and Technology    2016  [Refereed]

     View Summary

    Towards the actualization of an autonomous guitar teaching system, this paper proposes the following two video analysis based methods: (1) pressed chord recognition and (2) fingertip tracking. For (1), an algorithm that can extract finger contours and chord changes is proposed so that the chords pressed by the guitar player are recognized. For (2), an algorithm that can track the fingertips by continuously monitoring the appearance and disappearance of the regions of fingertip candidates is proposed. Experimental results demonstrate that the proposed two modules are robust enough under complex contexts such as complicated background and different illumination conditions. Promising results were obtained for accurate tracking of fingertips and for accurate recognition of pressed chords.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Detecting and Tracking the Guitar Neck Towards the Actualization of a Guitar Teaching-aid System

    Zhao Wang, Jun Ohya

    6th International Conference on Advanced Mechatronics (ICAM2015)   2015   187 - 188  2015.12  [Refereed]

     View Summary

    Towards the actualization of an autonomous guitar teaching-aid system, this paper proposes the following two video analysis based methods: (1) guitar neck detecting and (2) guitar neck tracking. For (1), an algorithm that can extract the rectangles cut by guitar strings and fret is proposed so that we can detect guitar neck area on the first frame of the input video. For (2), a feature point based guitar neck tracking algorithm is proposed, which can cope with occlusion issue effectively by analyzing geometric relationship between each two feature points in every frame. Experimental results demonstrate that the proposed two modules are robust enough under complex contexts such as complicated background and different illumination conditions.

    DOI CiNii

  • Estimating Clean-up Robots’ Mechanical Operations of Objects Using a SLAM Based Method

    Si Gao, Naotomo Tatematsu, Jun Ohya, Zhao Wang

    6th International Conference on Advanced Mechatronics (ICAM2015)   2015   249 - 250  2015.12  [Refereed]

     View Summary

    To enable clean-up robots to vacuum cluttered floors, the robots should be able to return objects on the floors to the proper positions by pushing the objects. This paper proposes a computer vision based method that can detect the objects (clutter) and evaluate the result of the mechanical operation (pushing the objects). First, we construct a dense 3D map of a tidy situation of the room (after-clean-map). Next, we construct another dense 3D map of the cluttered situation of the same room (before-clean-map). Then, we construct the 3D point cloud of the cluttered situation by comparing the after-clean-map and the before-clean-map. Finally, we let a clean-up robot push the cluttered object and evaluate whether the object is successfully moved. Experimental results using real RGBD image sequences demonstrate promising results of our proposed method.

    DOI CiNii

  • Early Recognition of Surgeons' Hand Actions from Continuous Surgery Videos

    Ye Li, Jun Ohya

    Interactive Medical Image Computing (IMIC), affiliated with MICCAI2015     78 - 85  2015.10  [Refereed]

  • Recognizing Hand Actions during Suture Surgeries from Video Sequences Based on Sliding Windows, SBoWS Feature and Neural Network

    Ye LI, Jun OHYA, Toshio CHIBA, Rong XU, Hiromasa YAMASHITA

    IIEEJ Transactions on Image Electronics and Visual Computing   3 ( 1 ) 34 - 43  2015.06  [Refereed]

    CiNii

  • Enhancing abstract imaginations of viewers of abstract paintings by a gaze based music generation system

    Tatsuya Ogusu, Jun Ohya, Jun Kurumisawa, Shunichi Yonemura

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9173 ( LNCS 9173 ) 457 - 464  2015  [Refereed]

     View Summary

    The purpose of abstract painters is to let viewers get the various images and abstract images. However, viewers who do not have enough knowledge of art, cannot easily get abstract images. The authors have proposed a music generation system that utilizes viewers’ gazes. It can be expected that the authors’ music generation system can prompt the viewer of abstract paintings to imagine abstract images, which the painter intended to express. This paper explores whether the authors’ music generation system can enhance abstract imaginations of persons who see abstract paintings, by subjective tests. Experiments using 19 subjects and eight abstract paintings were conducted for the two cases in which the subjects see the abstract paintings without hearing any music and while hearing the viewers’ gaze based music generated by the authors’ system. Experimental results imply that “hearing gaze based music” could enhance the viewers’ abstract imagination.

    DOI

    Scopus

  • Enhancing abstract imaginations of viewers of abstract paintings by a gaze based music generation system

    Tatsuya Ogusu, Jun Ohya, Jun Kurumisawa, Shunichi Yonemura

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   9173   457 - 464  2015  [Refereed]

     View Summary

    The purpose of abstract painters is to let viewers get the various images and abstract images. However, viewers who do not have enough knowledge of art, cannot easily get abstract images. The authors have proposed a music generation system that utilizes viewers’ gazes. It can be expected that the authors’ music generation system can prompt the viewer of abstract paintings to imagine abstract images, which the painter intended to express. This paper explores whether the authors’ music generation system can enhance abstract imaginations of persons who see abstract paintings, by subjective tests. Experiments using 19 subjects and eight abstract paintings were conducted for the two cases in which the subjects see the abstract paintings without hearing any music and while hearing the viewers’ gaze based music generated by the authors’ system. Experimental results imply that “hearing gaze based music” could enhance the viewers’ abstract imagination.

    DOI

    Scopus

  • Detection and segmentation of moving objects from dynamic RGB and depth images

    Naotomo Tatematsu, Jun Ohya, Larry Davis

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   8971   19 - 34  2015  [Refereed]

     View Summary

    This paper proposes a method that combines Temporal Modified-RANSAC(TMR) with a fixation-based segmentation algorithm for reconstructing the 3D structure of moving and still objects that are tracked in video and depth image sequences acquired by moving Kinect© and/or range finders First, we compute 3D optical flow of feature points. Second, TMR classifies all flows into consistent 3D flow sets for the background and each moving object
    simultaneously, the rotation matrix and the translation vector for each 3D flow set are computed. Then, an improved fixation-based segmentation method segments each object’s area. Finally, dense 3D models for the background and each moving object are constructed along with each object’s rotation matrix and translation vector in each frame. Experiments using multiple moving objects in color and depth image sequences acquired by Kinect(c) demonstrate the effectiveness of our proposed method.

    DOI

    Scopus

  • Automatic Neural Network Based Generation of Line Drawing Based Portraits - Study of Conditions for Accurate Portrait Generation and Representing Noses -

    Makoto ENOMOTO, Guoyong FANG, Yuzhen WU, Jun OHYA

    Fourth IIEEJ International Workshop on Image Electronics and Visual Computing   ( 2C-3 ) 1 - 4  2014.10  [Refereed]

  • Study of Recognizing Customers’ Behaviors Based on Facial Expression Analysis by a Bayesian Network

    Luwei JIA, Jun OHYA, Takayuki HORI

    Fourth IIEEJ International Workshop on Image Electronics and Visual Computing   ( 2P-8 ) 1 - 4  2014.10  [Refereed]

  • A Bayesian Network Based Classification of Customers’ Behaviors in Stores Using Image Processing Based Extraction of Behavior Features

    Jing Yan, Takayuki Hori, Jun Ohya

    Fourth IIEEJ International Workshop on Image Electronics and Visual Computing   ( 4A-4 ) 1 - 4  2014.10  [Refereed]

  • Locating Fetal Facial Surface, Oral Cavity and Airways by a 3D Ultrasound Calibration Using a Novel Cones' Phantom

    Rong Xu, Jun Ohya, Yoshinobu Sato, Bo Zhang, Masakatsu G. Fujie

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E97D ( 5 ) 1324 - 1335  2014.05  [Refereed]

     View Summary

    Toward the actualization of an automatic navigation system for fetoscopic tracheal occlusion (FETO) surgery, this paper proposes a 3D ultrasound (US) calibration-based approach that can locate the fetal facial surface, oral cavity, and airways by a registration between a 3D fetal model and 3D US images. The proposed approach consists of an offline process and online process. The offline process first reconstructs the 3D fetal model with the anatomies of the oral cavity and airways. Then, a point-based 3D US calibration system based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom, computes the matrix that transforms the 3D US image space into the world coordinate system. In the online process, by scanning the mother's body with a 3D US probe, 3D US images containing the fetus are obtained. The fetal facial surface extracted from the 3D US images is registered to the 3D fetal model using an ICP-based (iterative closest point) algorithm and the calibration matrices, so that the fetal facial surface as well as the oral cavity and airways are located. The results indicate that the 3D US calibration system achieves an FRE (fiducial registration error) of 1.49 +/- 0.44 mm and a TRE (target registration error) of 1.81 +/- 0.56 mm by using 24 fiducial points from two US volumes. A mean TRE of 1.55 +/- 0.46 mm is also achieved for measuring location accuracy of the 3D fetal facial surface extracted from 3D US images by 14 target markers, and mean location errors of 2.51 +/- 0.47 mm and 3.04 +/- 0.59 mm are achieved for indirectly measuring location accuracy of the pharynx and the entrance of the trachea, respectively, which satisfy the requirement of the FETO surgery.

    DOI

    Scopus

  • 相異なる照明光下の固定カメラ画像からの照明光色推定法

    川村春美, 八尾泰洋, 米村俊一, 大谷 淳, 小島 明

    画像電子学会誌   43 ( 2 ) 164 - 174  2014.03  [Refereed]

    DOI J-GLOBAL

  • Recognizing Surgeon's Actions during Suture Operations from Video Sequences

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    MEDICAL IMAGING 2014: IMAGE PROCESSING   9034  2014  [Refereed]

     View Summary

    Because of the shortage of nurses in the world, the realization of a robotic nurse that can support surgeries autonomously is very important. More specifically, the robotic nurse should be able to autonomously recognize different situations of surgeries so that the robotic nurse can pass necessary surgical tools to the medical doctors in a timely manner. This paper proposes and explores methods that can classify suture and tying actions during suture operations from the video sequence that observes the surgery scene that includes the surgeon's hands.
    First, the proposed method uses skin pixel detection and foreground extraction to detect the hand area. Then, interest points are randomly chosen from the hand area so that their 3D SIFT descriptors are computed. A word vocabulary is built by applying hierarchical K-means to these descriptors, and the words' frequency histogram, which corresponds to the feature space, is computed Finally, to classify the actions, either SVM (Support Vector Machine), Nearest Neighbor rule (NN) for the feature space or a method that combines "sliding window" with NN is performed.
    We collect 53 suture videos and 53 tying videos to build the training set and to test the proposed method experimentally It turns out that the NN gives higher than 90% accuracies, which are better recognition than SVM. Negative actions, which are different from either suture or tying action, are recognized with quite good accuracies, while "Sliding window" did not show significant improvements for suture and tying and cannot recognize negative actions.

    DOI

    Scopus

  • Automatic Recognition of Square Notation Symbols in Western Plainchant Manuscripts

    Carolina Ramirez, Jun Ohya

    JOURNAL OF NEW MUSIC RESEARCH   43 ( 4 ) 390 - 399  2014  [Refereed]

     View Summary

    While the Optical Music Recognition (OMR) of printed and handwritten music scores in modern standard notation has been broadly studied, this is not the case for early music manuscripts. This is mainly due to the high variability in the sources introduced by their severe physical degradation, the lack of notation standards and, in the case of the scanned versions, by non-homogenous image-acquisition protocols. The volume of early musical manuscripts available is considerable, and therefore we believe that computational methods can be extremely useful in helping to preserve, share and analyse this information. This paper presents an approach to recognizing handwritten square musical notation in degraded western plainchant manuscripts from the XIVth to XVIth centuries. We propose the use of image processing techniques that behave robustly under high data variability and which do not require strong hypotheses regarding the condition of the sources. The main differences from traditional OMR approaches are our avoidance of the staff line removal stage and the use of grey-level images to perform primitive segmentation and feature extraction. We used 136 images from the Digital Scriptorium repository (DS, 2007), from which we were able to extract over 90% of the staves and over 88% of all symbols present. For symbol classification, we used gradient-based features and SVM classifiers, obtaining over 90% precision and recall over eight basic symbol classes.

    DOI

    Scopus

    14
    Citation
    (Scopus)
  • Recognizing surgeon's actions during suture operations from video sequences

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    Progress in Biomedical Optics and Imaging - Proceedings of SPIE   9034  2014  [Refereed]

     View Summary

    Because of the shortage of nurses in the world, the realization of a robotic nurse that can support surgeries autonomously is very important. More specifically, the robotic nurse should be able to autonomously recognize different situations of surgeries so that the robotic nurse can pass necessary surgical tools to the medical doctors in a timely manner. This paper proposes and explores methods that can classify suture and tying actions during suture operations from the video sequence that observes the surgery scene that includes the surgeon's hands. First, the proposed method uses skin pixel detection and foreground extraction to detect the hand area. Then, interest points are randomly chosen from the hand area so that their 3D SIFT descriptors are computed. A word vocabulary is built by applying hierarchical K-means to these descriptors, and the words frequency histogram, which corresponds to the feature space, is computed. Finally, to classify the actions, either SVM (Support Vector Machine), Nearest Neighbor rule (NN) for the feature space or a method that combines sliding window with NN is performed. We collect 53 suture videos and 53 tying videos to build the training set and to test the proposed method experimentally. It turns out that the NN gives higher than 90% accuracies, which are better recognition than SVM. Negative actions, which are different from either suture or tying action, are recognized with quite good accuracies, while Sliding window did not show significant improvements for suture and tying and cannot recognize negative actions. © 2014 SPIE.

    DOI

    Scopus

  • Inspiring Viewers of Abstract Painting by a Gaze Based Music Generation

    Tatsuya Ogusu, Jun Ohya, Jun Kurumisawa, Shunichi Yonemura

    Communications in Computer and Information Science   434   213 - 217  2014  [Refereed]

     View Summary

    This paper explores the effectiveness of prompting abstract paintings' viewers' inspiration and imagination by the authors' gaze based music generation system. The authors' music generation system detects the viewer's gaze by a gaze detection equipment. At each of the gaze staying positions in the painting, the color of that point is converted to the sound so that as the gaze moves, music that consists of the converted time series sounds is generated. Experiments using six subjects and six abstract paintings were conducted for the three cases in which the subjects see the abstract paintings without hearing any music, while hearing pre-selected music and while hearing the viewers' gaze based music generated by the authors' system. The experimental results imply that "hearing gaze based music" could stimulate the viewers' inspiration and imagination best, "hearing pre-selected music" second best, and "without music" third best. © Springer International Publishing Switzerland 2014.

    DOI

    Scopus

  • Locating fetal facial surface, oral cavity and airways by a 3D ultrasound calibration using a novel cones' phantom

    Rong Xu, Jun Ohya, Yoshinobu Sato, Bo Zhang, Masakatsu G. Fujie

    IEICE Transactions on Information and Systems   E96-D ( 5 ) 1324 - 1335  2014  [Refereed]

     View Summary

    Toward the actualization of an automatic navigation system for fetoscopic tracheal occlusion (FETO) surgery, this paper proposes a 3D ultrasound (US) calibration-based approach that can locate the fetal facial surface, oral cavity, and airways by a registration between a 3D fetal model and 3D US images. The proposed approach consists of an offline process and online process. The offline process first reconstructs the 3D fetal model with the anatomies of the oral cavity and airways. Then, a point-based 3D US calibration system based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom, computes the matrix that transforms the 3D US image space into the world coordinate system. In the online process, by scanning the mother's body with a 3D US probe, 3D US images containing the fetus are obtained. The fetal facial surface extracted from the 3D US images is registered to the 3D fetal model using an ICP-based (iterative closest point) algorithm and the calibration matrices, so that the fetal facial surface as well as the oral cavity and airways are located. The results indicate that the 3D US calibration system achieves an FRE (fiducial registration error) of 1.49±0.44mm and a TRE (target registration error) of 1.81±0.56mm by using 24 fiducial points from two US volumes. A mean TRE of 1.55±0.46 mm is also achieved for measuring location accuracy of the 3D fetal facial surface extracted from 3D US images by 14 target markers, and mean location errors of 2.51±0.47 mm and 3.04±0.59 mm are achieved for indirectly measuring location accuracy of the pharynx and the entrance of the trachea, respectively, which satisfy the requirement of the FETO surgery. © 2014 The Institute of Electronics, Information and Communication Engineers.

    DOI

    Scopus

  • Automatic Lip Reading by Using Multimodal Visual Features

    Shohei Takahashi, Jun Ohya

    INTELLIGENT ROBOTS AND COMPUTER VISION XXXI: ALGORITHMS AND TECHNIQUES   9025  2014  [Refereed]

     View Summary

    Since long time ago, speech recognition has been researched, though it does not work well in noisy places such as in the car or in the train. In addition, people with hearing-impaired or difficulties in hearing cannot receive benefits from speech recognition. To recognize the speech automatically, visual information is also important. People understand speeches from not only audio information, but also visual information such as temporal changes in the lip shape. A vision based speech recognition method could work well in noisy places, and could be useful also for people with hearing disabilities.
    In this paper, we propose an automatic lip-reading method for recognizing the speech by using multimodal visual information without using any audio information such as speech recognition. First, the ASM (Active Shape Model) is used to track and detect the face and lip in a video sequence. Second, the shape, optical flow and spatial frequencies of the lip features are extracted from the lip detected by ASM. Next, the extracted multimodal features are ordered chronologically so that Support Vector Machine is performed in order to learn and classify the spoken words. Experiments for classifying several words show promising results of this proposed method.

    DOI

    Scopus

  • Quasi Random Scheme Based Improved Iterative Randomized Hough Transform (IRHT) for Detection of Partial Ellipses in Medical Images

    Rong XU, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    IIEEJ Transactions on Image Electronics and Visual Computing   1 ( 1 ) 97 - 106  2013.12  [Refereed]

    CiNii

  • 高彩度・低彩度gamut と反対色特性を利用した灰色仮説ベースの照明光色推定法

    川村春美, 米村俊一, 大谷 淳, 小島 明

    電子情報通信学会論文誌   J96-D ( 12 ) 3079 - 3089  2013.12  [Refereed]

    CiNii

  • Towards Recognizing Surgeon's Action during Suture Surgery from the Video Sequence

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    The 9th Asian Conference on Computer Aided Surgery     66 - 67  2013.09  [Refereed]

  • Automatic Real-time Tracking of Fetal Mouth in Fetoscopic Video Sequence for Supporting Fetal Surgeries

    Rong Xu, Tianliang Xie, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    MEDICAL IMAGING 2013: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING   8671   1 - 7  2013  [Refereed]

     View Summary

    Recently, a minimally invasive surgery (MIS) called fetoscopic tracheal occlusion (FETO) was developed to treat severe congenital diaphragmatic hernia (CDH) via fetoscopy, by which a detachable balloon is placed into the fetal trachea for preventing pulmonary hypoplasia through increasing the pressure of the chest cavity. This surgery is so dangerous that a supporting system for navigating surgeries is deemed necessary. In this paper, to guide a surgical tool to be inserted into the fetal trachea, an automatic approach is proposed to detect and track the fetal face and mouth via fetoscopic video sequencing. More specifically, the AdaBoost algorithm is utilized as a classifier to detect the fetal face based on Haar-like features, which calculate the difference between the sums of the pixel intensities in each adjacent region at a specific location in a detection window. Then, the CamShift algorithm based on an iterative search in a color histogram is applied to track the fetal face, and the fetal mouth is fitted by an ellipse detected via an improved iterative randomized Hough transform approach. The experimental results demonstrate that the proposed automatic approach can accurately detect and track the fetal face and mouth in real-time in a fetoscopic video sequence, as well as provide an effective and timely feedback to the robot control system of the surgical tool for FETO surgeries.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Gray-world-assumption-based illuminant color estimation using color gamuts with high and low chroma

    Harumi Kawamura, Shunichi Yonemura, Jun Ohya, Akira Kojima

    Proceedings of SPIE - The International Society for Optical Engineering   8652 ( 86520C ) 1 - 8  2013  [Refereed]

     View Summary

    A new approach is proposed for estimating illuminant colors from color images under an unknown scene illuminant. The approach is based on a combination of a gray-world-assumption-based illuminant color estimation method and a method using color gamuts. The former method, which is one we had previously proposed, improved on the original method that hypothesizes that the average of all the object colors in a scene is achromatic. Since the original method estimates scene illuminant colors by calculating the average of all the image pixel values, its estimations are incorrect when certain image colors are dominant. Our previous method improves on it by choosing several colors on the basis of an opponent-color property, which is that the average color of opponent colors is achromatic, instead of using all colors. However, it cannot estimate illuminant colors when there are only a few image colors or when the image colors are unevenly distributed in local areas in the color space. The approach we propose in this paper combines our previous method and one using high chroma and low chroma gamuts, which makes it possible to find colors that satisfy the gray world assumption. High chroma gamuts are used for adding appropriate colors to the original image and low chroma gamuts are used for narrowing down illuminant color possibilities. Experimental results obtained using actual images show that even if the image colors are localized in a certain area in the color space, the illuminant colors are accurately estimated, with smaller estimation error average than that generated in the conventional method. © 2013 SPIE.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • A flexible surgical tool localization using a 3D ultrasound calibration system for fetoscopic tracheal occlusion (FETO)

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   7761   17 - 24  2013  [Refereed]

     View Summary

    In fetoscopic tracheal occlusion (FETO) surgery, the localization of a flexible tool has an important role in easing high demands for surgeons. In this paper, a point-based 3D ultrasound (US) calibration system is proposed for localizing the surgical tool, based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom. A 3D US probe attached with an EM sensor is used to acquire the 3D US images of the surgical tool
    meanwhile, an EM transmitter is fixed at a certain point in the environment as the origin of the world coordinate system. From these 3D US images, the skeleton of the surgical tool is extracted, and 15 labeled points along the surgical tool are then distinguished, whose coordinates in the world coordinate system are calculated by the proposed 3D US calibration system. The results demonstrate that our proposed approach achieves high localization accuracy, i.e. the average TRE of 1.99 ± 0.49 mm and the maximum TRE of 2.61 ± 0.49 mm, which satisfy the requirement of the FETO surgery: less than 3.85 mm. © Springer-Verlag 2013.

    DOI

    Scopus

  • Automatic Fetal Face Detection By Locating Fetal Facial Features From 3D Ultrasound Images For Navigating Fetoscopic Tracheal Occlusion Surgeries

    Rong Xu, Jun Ohya, Bo Zhang, Masakatsu G. Fujie, Yoshinobu Sato

    2013 10TH IEEE INTERNATIONAL CONFERENCE AND WORKSHOPS ON AUTOMATIC FACE AND GESTURE RECOGNITION (FG)   ( 104 ) 1 - 6  2013  [Refereed]

     View Summary

    With the wide clinical application of 3D ultrasound (US) imaging, automatic location of fetal facial features from US volumes for navigating fetoscopic tracheal occlusion (FETO) surgeries becomes possible, which plays an important role in reducing surgical risk. In this paper, we propose a feature-based method to automatically detect 3D fetal face and accurately locate key facial features without any priori knowledge or training data. The candidates of the key facial features, such as the nose, eyes, nose upper bridge and upper lip are detected by analyzing the mean and Gaussian curvatures of the facial surface. Each feature is gradually identified from the candidates by a boosting traversal scheme based on the spatial relations between each feature. In experiments, all key feature points are detected for each case, and thus a detection success rate of 100% is achieved by using 72 3D US images from a test database of 6 fetal faces in the frontal view and any pose within 15 degrees from the frontal view, and the location error 3.18 +/- 0.91 inn of the detected upper lip for all test data is obtained, which can be tolerated by the FETO surgery. Moreover, this system has a high efficiency and can detect all key facial features in about 625 ms on a quadcore 2.60 GHz computer.

  • An investigation into encouraging higher user self-disclosure during text-based computer mediated communication: A case study of prompter communication system

    Li Jen Chen, Mutsumi Suganuma, Shigekazu Sakai, Jun Ohya, Shunichi Yonemura, Yukio Tokunaga

    Journal of the Institute of Image Electronics Engineers of Japan   42 ( 1 ) 89 - 101  2013.01  [Refereed]

     View Summary

    An investigation into encouraging higher user self-disclosure during text-based Computer Mediated Communication (CMC) system known as the Prompter Communication system (PCS), proposed by Yonemura et al. in previous studies, in which 1 to 3 black circular prompter images ("•") are embedded onto the background of a text-field canvas as stimuli and as an embryonic basis for conversation, have been investigated and evaluated. The effects on supporting the projection or reflection of generated mental representations onto messages with the "•" prompter image as part of the representations within the message contents have also been examined. During the investigation, the concepts and features of the Geneplore cognitive creative thinking process model, together with self-projections, were integrated into the proposed system. Based on these concepts and features, the PCS aims to encourage users to express their mental representations by projecting or reflecting them onto the messages in combination with the provided "•" prompter image as part of the message contents. Communication experiments using the prototype PCS system have been conducted, where a Normal Communication System, NCS which uses only a plain text-field canvas without the "•" prompter image, is compared with the PCS. The experimental results indicate that communication conducted using the PCS results in an increased number of messages with the "•" prompter image(s) involved as part of the message contents than as background images. Further analysis into the self-disclosure level of the message contents and the relationships between the degree of self-disclosure and the representations of the "•" prompter image(s) within the messages have indicated that the degrees of self-disclosure increases when the "•" prompter image(s) are used to represent the participant him/herself, or the participant's thoughts and emotions, and decreases when the "•" prompter image(s) are used to represent animals or objects, or when the "•" prompter image(s) presented within the messages have been ignored. Investigations into the differences in self-disclosure level between message contents from both the PCS and the NCS have shown that using the PCS results in a generally higher degree in the average frequency of the appearance of high self-disclosure information than when using the NCS. During the conversation, these messages contain information with a high level of self-disclosure, which, consequently, may further lead to the development of intimate and deep interpersonal relationships.

    DOI

  • A flexible surgical tool localization using a 3D ultrasound calibration system for fetoscopic tracheal occlusion (FETO)

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   7761   17 - 24  2013  [Refereed]

     View Summary

    In fetoscopic tracheal occlusion (FETO) surgery, the localization of a flexible tool has an important role in easing high demands for surgeons. In this paper, a point-based 3D ultrasound (US) calibration system is proposed for localizing the surgical tool, based on real-time 3D US images, an electromagnetic (EM) tracking device, and a novel cones' phantom. A 3D US probe attached with an EM sensor is used to acquire the 3D US images of the surgical tool
    meanwhile, an EM transmitter is fixed at a certain point in the environment as the origin of the world coordinate system. From these 3D US images, the skeleton of the surgical tool is extracted, and 15 labeled points along the surgical tool are then distinguished, whose coordinates in the world coordinate system are calculated by the proposed 3D US calibration system. The results demonstrate that our proposed approach achieves high localization accuracy, i.e. the average TRE of 1.99 ± 0.49 mm and the maximum TRE of 2.61 ± 0.49 mm, which satisfy the requirement of the FETO surgery: less than 3.85 mm. © Springer-Verlag 2013.

    DOI

    Scopus

  • Gray-world-assumption-based illuminant color estimation using color gamuts with high and low chroma

    Harumi Kawamura, Shunichi Yonemura, Jun Ohya, Akira Kojima

    Proceedings of SPIE - The International Society for Optical Engineering   8652  2013  [Refereed]

     View Summary

    A new approach is proposed for estimating illuminant colors from color images under an unknown scene illuminant. The approach is based on a combination of a gray-world-assumption-based illuminant color estimation method and a method using color gamuts. The former method, which is one we had previously proposed, improved on the original method that hypothesizes that the average of all the object colors in a scene is achromatic. Since the original method estimates scene illuminant colors by calculating the average of all the image pixel values, its estimations are incorrect when certain image colors are dominant. Our previous method improves on it by choosing several colors on the basis of an opponent-color property, which is that the average color of opponent colors is achromatic, instead of using all colors. However, it cannot estimate illuminant colors when there are only a few image colors or when the image colors are unevenly distributed in local areas in the color space. The approach we propose in this paper combines our previous method and one using high chroma and low chroma gamuts, which makes it possible to find colors that satisfy the gray world assumption. High chroma gamuts are used for adding appropriate colors to the original image and low chroma gamuts are used for narrowing down illuminant color possibilities. Experimental results obtained using actual images show that even if the image colors are localized in a certain area in the color space, the illuminant colors are accurately estimated, with smaller estimation error average than that generated in the conventional method. © 2013 SPIE.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Automatic Real-time Tracking of Fetal Mouth in Fetoscopic Video Sequence for Supporting Fetal Surgeries

    Rong Xu, Tianliang Xie, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    MEDICAL IMAGING 2013: IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING   8671  2013  [Refereed]

     View Summary

    Recently, a minimally invasive surgery (MIS) called fetoscopic tracheal occlusion (FETO) was developed to treat severe congenital diaphragmatic hernia (CDH) via fetoscopy, by which a detachable balloon is placed into the fetal trachea for preventing pulmonary hypoplasia through increasing the pressure of the chest cavity. This surgery is so dangerous that a supporting system for navigating surgeries is deemed necessary. In this paper, to guide a surgical tool to be inserted into the fetal trachea, an automatic approach is proposed to detect and track the fetal face and mouth via fetoscopic video sequencing. More specifically, the AdaBoost algorithm is utilized as a classifier to detect the fetal face based on Haar-like features, which calculate the difference between the sums of the pixel intensities in each adjacent region at a specific location in a detection window. Then, the CamShift algorithm based on an iterative search in a color histogram is applied to track the fetal face, and the fetal mouth is fitted by an ellipse detected via an improved iterative randomized Hough transform approach. The experimental results demonstrate that the proposed automatic approach can accurately detect and track the fetal face and mouth in real-time in a fetoscopic video sequence, as well as provide an effective and timely feedback to the robot control system of the surgical tool for FETO surgeries.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Automatic Fetal Face Detection By Locating Fetal Facial Features From 3D Ultrasound Images For Navigating Fetoscopic Tracheal Occlusion Surgeries

    Rong Xu, Jun Ohya, Bo Zhang, Masakatsu G. Fujie, Yoshinobu Sato

    2013 10TH IEEE INTERNATIONAL CONFERENCE AND WORKSHOPS ON AUTOMATIC FACE AND GESTURE RECOGNITION (FG)    2013  [Refereed]

     View Summary

    With the wide clinical application of 3D ultrasound (US) imaging, automatic location of fetal facial features from US volumes for navigating fetoscopic tracheal occlusion (FETO) surgeries becomes possible, which plays an important role in reducing surgical risk. In this paper, we propose a feature-based method to automatically detect 3D fetal face and accurately locate key facial features without any priori knowledge or training data. The candidates of the key facial features, such as the nose, eyes, nose upper bridge and upper lip are detected by analyzing the mean and Gaussian curvatures of the facial surface. Each feature is gradually identified from the candidates by a boosting traversal scheme based on the spatial relations between each feature. In experiments, all key feature points are detected for each case, and thus a detection success rate of 100% is achieved by using 72 3D US images from a test database of 6 fetal faces in the frontal view and any pose within 15 degrees from the frontal view, and the location error 3.18 +/- 0.91 inn of the detected upper lip for all test data is obtained, which can be tolerated by the FETO surgery. Moreover, this system has a high efficiency and can detect all key facial features in about 625 ms on a quadcore 2.60 GHz computer.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • 3D Surface Reconstruction Based on Image Stitching from Gastric Endoscopic Video Sequence

    Mengyao Duan, Rong Xu, Jun Ohya

    APPLICATIONS OF DIGITAL IMAGE PROCESSING XXXVI   8856  2013  [Refereed]

     View Summary

    This paper proposes a method for reconstructing 3D detailed structures of internal organs such as gastric wall from endoscopic video sequences. The proposed method consists of the four major steps: Feature-point-based 3D reconstruction, 3D point cloud stitching, dense point cloud creation and Poisson surface reconstruction. Before the first step, we partition one video sequence into groups, where each group consists of two successive frames (image pairs), and each pair in each group contains one overlapping part, which is used as a stitching region. Fist, the 3D point cloud of each group is reconstructed by utilizing structure from motion (SFM). Secondly, a scheme based on SIFT features registers and stitches the obtained 3D point clouds, by estimating the transformation matrix of the overlapping part between different groups with high accuracy and efficiency. Thirdly, we select the most robust SIFT feature points as the seed points, and then obtain the dense point cloud from sparse point cloud via a depth testing method presented by Furukawa. Finally, by utilizing Poisson surface reconstruction, polygonal patches for the internal organs are obtained. Experimental results demonstrate that the proposed method achieves a high accuracy and efficiency for 3D reconstruction of gastric surface from an endoscopic video sequence.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • e-learning 用講義ビデオにおける 学習者の学習行動を利用した主観的難易度の把握方法の基礎的検討

    大川内隆朗, 大谷淳, 米村俊一, 徳永幸生

    日本教育工学会論文誌   36 ( 3 ) 193 - 203  2012.12  [Refereed]

    DOI J-GLOBAL

  • Study of a teacher support method for improving lectures for e-learning on demand

    Takaaki Ohkawauchi, Jun Ohya, Shunichi Yonemura, Yukio Tokunaga

    IEVC (IIEEJ 3rd Image Electronics and Visual Computing Workshop) 2012   ( 2A-4 ) 1 - 4  2012.11  [Refereed]

  • Study of generating portraits using neural networks - Portraits that consist of facial contour, eyebrows, eyes and mouth ?

    Makoto Enomoto, Akira Takano, Jun Ohya

    IEVC (IIEEJ 3rd Image Electronics and Visual Computing Workshop) 2012   ( 4A-4 ) 1 - 4  2012.11  [Refereed]

  • Proposed Prompter Communication System for Supporting Non-verbal Expressions in order to Encourage Development of Deep and Intimate Interpersonal Relationships

    Chen Li Jen, Mutsumi Suganuma, Shigekazu Sakai, Shunichi Yonemura, Jun Ohya, Yukio Tokunaga

    早稲田大学GITS/GITI紀要2011-2012     20 - 30  2012.10  [Refereed]

    CiNii

  • e-learning用講義コンテンツにおける教員の講義改善活動を支援するシステムの提案と評価

    大川内隆朗, 大谷淳, 米村俊一, 徳永幸生

    早稲田大学GITS/GITI紀要   2011-2012   11 - 19  2012.10  [Refereed]

    J-GLOBAL

  • Detection and Segmentation of Moving Objects from Dynamic RGB and Depth Images

    Naotomo Tatematsu, Jun Ohya, Larry Davis

    3D Camera Workshop in conjunction with International Conference on Virtual Reality and Visualization     1 - 8  2012.09  [Refereed]

  • Estimating object colors in outdoor scenes with same object region but different illuminants

    Harumi Kawamura, Yasuhiro Yao, Shunichi Yonemura, Jun Ohya, Akira Kojima

    AIC (The International Colour Association) 2012 Interim Meeting     582 - 585  2012.09  [Refereed]

  • Generating Various Face Makeup Styles based on a Small Set of Examples

    Jieun Cho, Minje Park, Sang Il Park, Jun Ohya

    CGI (Computer Graphics International) 2012   ( 222 ) 1 - 4  2012.06  [Refereed]

  • Segmentation of Brain MRI

    Rong Xu, Limin Luo, Jun Ohya

    Advances in Brain Imaging     143 - 170  2012.02

  • Tracking White Road Line by Particle Filter from the Video Sequence Acquired by the Camera Attached to a Walking Human Body

    Shohei Takahashi, Jun Ohya

    IMAGE PROCESSING: ALGORITHMS AND SYSTEMS X AND PARALLEL PROCESSING FOR IMAGING APPLICATIONS II   8295   82950 - 82959  2012  [Refereed]

     View Summary

    This paper proposes a method for tracking and recognizing the white line marked in the surface of the road from the video sequence acquired by the camera attached to a walking human, towards the actualization of an automatic navigation system for the visually handicapped. Our proposed method consists of two main modules: (1) Particle Filter based module for tracking the white line, and (2) CLAFIC Method based module for classifying whether the tracked object is the white line. In (1), each particle is a rectangle, and is described by its centroid's coordinates and its orientation. The likelihood of a particle is computed based on the number of white pixels in the rectangle. In (2), in order to obtain the ranges (to be used for the recognition) for the white line's length and width, Principal Component Analysis is applied to the covariance matrix obtained from valid sample particles. At each frame, PCA is applied to the covariance matrix constructed from particles with high likelihood, and if the obtained length and width are within the above-mentioned ranges, it is recognized as the white line. Experimental results using real video sequences show the validity of the proposed method.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Accurate, Dense 3D Reconstruction of Moving and Still Objects from Dynamic Stereo Sequences Based on Temporal Modified-RANSAC and Feature-Cut

    Naotomo Tatematsu, Jun Ohya

    INTELLIGENT ROBOTS AND COMPUTER VISION XXIX: ALGORITHMS AND TECHNIQUES   8301   830105 - 830115  2012  [Refereed]

     View Summary

    This paper improves the authors' conventional method for reconstructing the 3D structure of moving and still objects that are tracked in the video and/or depth image sequences acquired by moving cameras and/or range finder. The authors proposed a Temporal Modified-RANSAC based method [1] that (1) can discriminate each moving object from the still background in color image and depth image sequences acquired by moving stereo cameras or moving range finder, (2) can compute the stereo cameras' egomotion, (3) can compute the motion of each moving object, and (4) can reconstruct the 3D structure of each moving object and the background. However, the TMR based method has the following two problems concerning the 3D reconstruction: lack of accuracy of segmenting into each object's region and sparse 3D reconstructed points in each object's region. To solve these problems of our conventional method, this paper proposes a new 3D segmentation method that utilizes Graph-cut, which is frequently used for segmentation tasks. First, the proposed method tracks feature points in the color and depth image sequences so that 3D optical flows of the feature points in every N frames are obtained. Then, TMR classifies all the obtained 3D optical flows into regions (3D flow set) for the background and each moving object; simultaneously, the rotation matrix and the translation vector for each 3D flow set are computed. Next, Graph-Cut using the energy function that consists of color probability, structure probability and a-priori probability is performed so that pixels in each frame are segmented into object regions and the background region. Finally, 3D point clouds are obtained from the segmentation result image and depth image, and then the point clouds are merged using the rotation and translation from the N-th frame prior to the current frame so that 3D models for the background and each moving object are constructed with dense 3D point data.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Human Gesture Analysis using Multimodal features

    Luo Dan, Hazim Kemal Ekenel, Ohya Jun

    2012 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO WORKSHOPS (ICMEW)     471 - 476  2012  [Refereed]

     View Summary

    Human gesture as a natural interface plays an utmost important role for achieving intelligent Human Computer Interaction (HCI). Human gestures include different components of visual actions such as motion of hands, facial expression, and torso, to convey meaning. So far, in the field of gesture recognition, most previous works have focused on the manual component of gestures. In this paper, we present an appearance-based multimodal gesture recognition framework, which combines the different groups of features such as facial expression features and hand motion features which are extracted from image frames captured by a single web camera. We refer 12 classes of human gestures with facial expression including neutral, negative and positive meanings from American Sign Languages (ASL). We combine the features in two levels by employing two fusion strategies. At the feature level, an early feature combination can be performed by concatenating and weighting different feature groups, and PLS is used to choose the most discriminative elements by projecting the feature on a discriminative expression space. The second strategy is applied on decision level. Weighted decisions from single modalities are fused in a later stage. A condensation-based algorithm is adopted for classification. We collected a data set with three to seven recording sessions and conducted experiments with the combination techniques. Experimental results showed that facial analysis improve hand gesture recognition, decision level fusion performs better than feature level fusion.

    DOI

    Scopus

    6
    Citation
    (Scopus)
  • Human Gesture Analysis using Multimodal features

    Luo Dan, Hazim Kemal Ekenel, Ohya Jun

    2012 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO WORKSHOPS (ICMEW)     471 - 476  2012  [Refereed]

     View Summary

    Human gesture as a natural interface plays an utmost important role for achieving intelligent Human Computer Interaction (HCI). Human gestures include different components of visual actions such as motion of hands, facial expression, and torso, to convey meaning. So far, in the field of gesture recognition, most previous works have focused on the manual component of gestures. In this paper, we present an appearance-based multimodal gesture recognition framework, which combines the different groups of features such as facial expression features and hand motion features which are extracted from image frames captured by a single web camera. We refer 12 classes of human gestures with facial expression including neutral, negative and positive meanings from American Sign Languages (ASL). We combine the features in two levels by employing two fusion strategies. At the feature level, an early feature combination can be performed by concatenating and weighting different feature groups, and PLS is used to choose the most discriminative elements by projecting the feature on a discriminative expression space. The second strategy is applied on decision level. Weighted decisions from single modalities are fused in a later stage. A condensation-based algorithm is adopted for classification. We collected a data set with three to seven recording sessions and conducted experiments with the combination techniques. Experimental results showed that facial analysis improve hand gesture recognition, decision level fusion performs better than feature level fusion.

    DOI

    Scopus

    6
    Citation
    (Scopus)
  • Automatic Fetal Head Detection on Ultrasound Images by An Improved Iterative Randomized Hough Transform

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    Proceedings of 26th International Conference of Image and Vision Computing New Zealand (IVCNZ’ 2011)     288 - 292  2011.11  [Refereed]

  • Visualizing Large Procedural Volumetric Terrains Using Nested Clip-Boxes

    Sven Forstmann, Jun Ohya

    GITS/GITI 紀要     51 - 61  2011.10  [Refereed]

    CiNii

  • 反対色の特性を利用した色選択による灰色仮説判定に基づくカラー画像からの照明光推定法

    川村春美, 米村俊一, 大谷 淳, 松浦宣彦

    電子情報通信学会論文誌 D   J94-D ( 8 ) 1346 - 1358  2011.08  [Refereed]

    J-GLOBAL

  • Facial Analysis Aided Human Gesture Recognition for Human Computer Interaction

    Dan Luo, Hua Gao, Hazim Kemal Ekenel, Jun Ohya

    Proceedings of MVA2011 (IAPR Conference on Machine Vision Applications)     446 - 449  2011.06  [Refereed]

  • Conversation strategies and characteristics in line drawing video communication system

    Shun-Ichi Yonemurat, Ken Hanada, Yukio Tokunaga, Satoshi Shimada, Akira Suzuki, Jun Ohya

    Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers   65 ( 4 ) 515 - 522  2011.04  [Refereed]

     View Summary

    The characteristics of line drawing video communication are described in this paper. In the experiment, an original video and a line drawing video were compared. Subjects performed the same task using both video systems and their performances were assessed in terms of the number of utterances in the conversation and the subject's switching pause. The subject's conversation strategy was extracted based on gaze measurement. In the case of the subujects who often gaze at a partner's eyes during conversation, the switching pause for the line drawing video was significantly longer than for the original video. In both video formats, subjects who does not gaze at a partner's eyes had more switching pauses and utterances than the subjects who do.

    DOI

    Scopus

  • Illuminant Color Estimation by Hue Categorization Based on Gray World Assumption

    Harumi Kawamura, Shunichi Yonemura, Jun Ohya, Norihiko Matsuura

    COMPUTATIONAL IMAGING IX   7873 ( 787312 ) 1 - 12  2011  [Refereed]

     View Summary

    This paper proposes a gray world assumption based method for estimating an illuminant color from an image by hue categorization. The gray world assumption hypothesizes that the average color of all the objects in a scene is gray. However, it is difficult to estimate an illuminant color correctly if the colors of the objects in a scene are dominated by certain colors. To solve this problem, our method uses the opponent color properties that the average of a pair of opponent colors is gray. Thus our method roughly categorizes the colors derived from the image based on hue and selects them one by one from the hue categories until selected colors satisfy the gray world assumption. In our experiments, we used three kinds of illuminants (i.e., CIE standard illuminants A and D-65, and a fluorescent light) and two kinds of data sets. One data set satisfies the gray world assumption, and the other does not. Experiment results show that estimated illuminants are closer to the correct ones than those obtained with the conventional method and the estimation error for both using CIE standard illuminants A and D-65 by our method are within the barely noticeable difference in human color perception.

    DOI

    Scopus

    7
    Citation
    (Scopus)
  • OMR of Early Plainchant Manuscripts in Square Notation: a Two-stage System

    Carolina Ramirez, Jun Ohya

    DOCUMENT RECOGNITION AND RETRIEVAL XVIII   7874 ( 787417 ) 1 - 8  2011  [Refereed]

     View Summary

    While Optical Music Recognition (OMR) of modern printed and handwritten documents is considered a solved problem, with many commercial systems available today, the OMR of ancient musical manuscripts still remains an open problem. In this paper we present a system for the OMR of degraded western plainchant manuscripts in square notation from the XIV to XVI centuries. The system has two main blocks, the first one deals with symbol extraction and recognition, while the second one acts as an error detection stage for the first block outputs. For symbol extraction we use widely known image-processing techniques, such as Sobel filtering and Hough Transform, and SVM for classification. The error detection stage is implemented with a hidden Markov model (HMM), which takes advantage of a priori knowledge for this specific kind of music.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Appearance-based Human Gesture Recognition using Multimodal Features for Human Computer Interaction

    Dan Luo, Hua Gao, Hazim Kemal Ekenel, Jun Ohya

    HUMAN VISION AND ELECTRONIC IMAGING XVI   7865 ( 786509 ) 1 - 6  2011  [Refereed]

     View Summary

    The use of gesture as a natural interface plays an utmost important role for achieving intelligent Human Computer Interaction (HCI). Human gestures include different components of visual actions such as motion of hands, facial expression, and torso, to convey meaning. So far, in the field of gesture recognition, most previous works have focused on the manual component of gestures. In this paper, we present an appearance-based multimodal gesture recognition framework, which combines the different groups of features such as facial expression features and hand motion features which are extracted from image frames captured by a single web camera. We refer 12 classes of human gestures with facial expression including neutral, negative and positive meanings from American Sign Languages (ASL). We combine the features in two levels by employing two fusion strategies. At the feature level, an early feature combination can be performed by concatenating and weighting different feature groups, and LDA is used to choose the most discriminative elements by projecting the feature on a discriminative expression space. The second strategy is applied on decision level. Weighted decisions from single modalities are fused in a later stage. A condensation-based algorithm is adopted for classification. We collected a data set with three to seven recording sessions and conducted experiments with the combination techniques. Experimental results showed that facial analysis improve hand gesture recognition, decision level fusion performs better than feature level fusion.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Study of Recognizing Human Motion Observed from an Arbitrary Viewpoint Based on Decomposition of a Tensor Containing Multiple View Motions

    Takayuki Hori, Jun Ohya, Jun Kurumisawa

    COMPUTATIONAL IMAGING IX   7873 ( 787310 ) 1 - 10  2011  [Refereed]

     View Summary

    We propose a Tensor Decomposition based algorithm that recognizes the observed action performed by an unknown person and unknown viewpoint not included in the database. Our previous research aimed motion recognition from one single viewpoint. In this paper, we extend our approach for human motion recognition from an arbitrary viewpoint. To achieve this issue, we set tensor database which are multi-dimensional vectors with dimensions corresponding to human models, viewpoint angles, and action classes. The value of a tensor for a given combination of human silhouette model, viewpoint angle, and action class is the series of mesh feature vectors calculated each frame sequence. To recognize human motion, the actions of one of the persons in the tensor are replaced by the synthesized actions. Then, the core tensor for the replaced tensor is computed. This process is repeated for each combination of action, person, and viewpoint. For each iteration, the difference between the replaced and original core tensors is computed. The assumption that gives the minimal difference is the action recognition result. The recognition results show the validity of our proposed method, the method is experimentally compared with Nearest Neighbor rule. Our proposed method is very stable as each action was recognized with over 75% accuracy.

    DOI

    Scopus

  • Study of Temporal Modified-RANSAC Based Method for the Extraction and 3D Shape Reconstruction of Moving Objects from Dynamic Stereo Images and for Estimating the Camera Pose

    Naotomo Tatematsu, Jun Ohya

    INTELLIGENT ROBOTS AND COMPUTER VISION XXVIII: ALGORITHMS AND TECHNIQUES   7878 ( 78780E ) 1 - 13  2011  [Refereed]

     View Summary

    This paper proposes a Temporal Modified-RANSAC based method that can discriminate each moving object from the still background in the stereo video sequences acquired by moving stereo cameras, can compute the stereo cameras' egomotion, and can reconstruct the 3D structure of each moving object and the background. We compute 3D optical flows from the depth map and results of tracking feature points. We define "3D flow region" as a set of connected pixels whose 3D optical flows have a common rotation matrix and translation vector. Our Temporal Modified-RANSAC segments the detected 3D optical flows into 3D flow regions and computes the rotation matrix and translation vector for each 3D flow region. As opposed to the conventional Modified-RANSAC for only two frames, The Temporal Modified-RANSAC can handle temporal images with arbitrary length by performing the Modified-RANSAC to the set of a 3D flow region that classified in the latest frame and new 3D optical flows detected in the current frame iteratively. Finally, the 3D points computed from the depth map in all the frames are registered using each 3D flow region's matrix to the initial positions in the initial frame so that the 3D structures of the moving objects and still background are reconstructed. Experiments using multiple moving objects and real stereo sequences demonstrate promising results of our proposed method.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Efficient, High-Quality, GPU-Based Visualization of Voxelized Surface Data with Fine and Complicated Structures

    Sven Forstmann, Jun Ohya

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E93D ( 11 ) 3088 - 3099  2010.11  [Refereed]

     View Summary

    This paper proposes a GPU based method that can visualize voxelized surface data with fine and complicated features has high rendering quality at interactive frame rates and provides low memory consumption The surface data is compressed using run length encoding (RLE) for each level of detail (LOD) Then the loop for the rendering process is performed on the GPU for the position of the viewpoint at each time instant The scene is raycasted in planes where each plane is perpendicular to the horizontal plane in the world coordinate system and passes through the viewpoint For each plane one ray is cast to rasterize all RLE elements intersecting this plane starting from the viewpoint and ranging up to the maximum view distance This rasteritization process projects each RLE element passing the occlusion test onto the screen at a LOD that decreases with the distance of the RLE element from the viewpoint Finally the smoothing of voxels in screen space and full screen anti aliasing is performed To provide lighting calculations without storing the normal vector inside the RLE data structure our algorithm recovers the normal vectors from the rendered scene s depth butter After the viewpoint changes the same process is re executed for the new viewpoint Experiments using different scenes have shown that the proposed algorithm is faster than the equivalent CPU implementation and other related methods Our experiments further prove that this method is memory efficient and achieves high quality results.

    DOI

    Scopus

    5
    Citation
    (Scopus)
  • Studies on Enterprises’ Adaptive Strategy to Economic Responsibility under EPR

    Yiping Zhao, Jun Ohya

    Proceedings of 2nd International Conference on Sustainable Enterprises of the Future     254 - 257  2010.09  [Refereed]

  • 色相に着目した灰色仮説に基づく照明光推定法の一提案

    川村春美, 米村俊一, 大谷淳, 松浦宣彦

    情報処理学会研究報告   2010-AVM-70 ( 6 ) 1 - 6  2010.08

    J-GLOBAL

  • Symbol Classification Approach for OMR of Square Notation Manuscripts

    Carolina Ramirez, Jun Ohya

    Proceedings of the 11th International Society for Music Information Retrieval Conference (ISMIR 2010)     549 - 554  2010.08  [Refereed]

  • Study on Adaptive Management Model and Key Factors under Extended Producer Responsibility

    Yiping Zhao, Jun Ohya

    GITS/GITI 紀要 2009-2010 (早稲田大学大学院国際情報通信研究科 国際情報通信研究センター), 査読論文   2009   136 - 144  2010.07  [Refereed]

    CiNii

  • Elliptical Object Detection by a Modified RANSAC with Sampling Constraint from Boundary Curves' Clustering

    Yingdi Xie, Jun Ohya

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E93D ( 3 ) 611 - 623  2010.03  [Refereed]

     View Summary

    This paper proposes a method for detecting ellipses from an image despite (1) multiple colors within the ellipses, (2) partially occluded ellipses' boundaries, (3) noisy. locally deformed boundaries of ellipses, (4) presence of multiple objects other than the ellipses in the image, and (5) combinations of (1) through (4). After boundary curves are obtained by edge detection, by utilizing the first-order difference curves of the edge orientation of each pixel in the boundary curves, a segment-reconnect method obtains boundary clusters. Then, a modified RANSAC detects ellipses by choosing five pixels randomly from the boundary clusters, where overlapped ellipses are merged. Experimental results using synthesized images and real images demonstrate the effectiveness of the proposed method together with comparison with the Randomized Hough Transform, a well-known conventional method.

    DOI

    Scopus

    5
    Citation
    (Scopus)
  • Study of Recognizing Multiple Persons' Complicated Hand Gestures from the Video Sequence Acquired by a Moving Camera

    Luo Dan, Jun Ohya

    HUMAN VISION AND ELECTRONIC IMAGING XV   7527  2010  [Refereed]

     View Summary

    Recognizing hand gestures from the video sequence acquired by a dynamic camera could be a useful interface between humans and mobile robots. We develop a state based approach to extract and recognize hand gestures from moving camera images. We improved Human-Following Local Coordinate (HFLC) System, a very simple and stable method for extracting hand motion trajectories, which is obtained from the located human face, body part and hand blob changing factor. Condensation algorithm and PCA-based algorithm was performed to recognize extracted hand trajectories. In last research, this Condensation Algorithm based method only applied for one person's hand gestures. In this paper, we propose a principal component analysis (PCA) based approach to improve the recognition accuracy. For further improvement, temporal changes in the observed hand area changing factor are utilized as new image features to be stored in the database after being analyzed by PCA. Every hand gesture trajectory in the database is classified into either one hand gesture categories, two hand gesture categories, or temporal changes in hand blob changes. We demonstrate the effectiveness of the proposed method by conducting experiments on 45 kinds of sign language based Japanese and American Sign Language gestures obtained from 5 people. Our experimental recognition results show better performance is obtained by PCA based approach than the Condensation algorithm based method.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Study on Incentive Management of the End-of-life Products Collection and Recycling under EPR

    Yiping Zhao, Jun Ohya

    PROCEEDINGS OF 2010 INTERNATIONAL CONFERENCE ON ENVIRONMENTAL SCIENCE AND DEVELOPMENT     157 - 162  2010  [Refereed]

     View Summary

    In this paper, the end-of-life products (ELPs) collecting and recycling system is studied under the application of extended producer responsibility (EPR) programs. Economic performance is evaluated based on the decomposition and analysis of major cost and revenues components during the recycling process. We identified two problematic scenarios where EPR programs cannot run smoothly. The according prosperous incentive management strategies are discussed as the corresponding solution.

  • Integrated System for End-of-life Products Collecting and Recycling under Extended Producer Responsibility

    Yiping Zhao, Jun Ohya

    The Sixth International Conference on Technology, Knowledge and Society, Berlin     1 - 7  2010.01  [Refereed]

  • PCA-based Method for Recognizing Multiple Persons’ Hand Gestures from the Video Sequence Acquired by a Moving Camera

    Luo Dan, Jun Ohya

    SPIE-IS&T, Electronic Imaging2010   7527   75271N  2010.01  [Refereed]

  • Identifying a Walking Human by a Tensor Decomposition Based Approach and Tracking the Human across Discontinuous Fields of Views of Multiple Cameras

    Takayuki Hori, Jun Ohya, Jun Kurumisawa

    COMPUTATIONAL IMAGING VIII   7533   75330X  2010  [Refereed]

     View Summary

    This paper proposes a method that identifies and tracks a walking human across discontinuous fields of views of multiple cameras for the purpose of video surveillance. A typical video surveillance system has multiple cameras, but there are several spaces within the surveillance area that are not within any of the camera's field of view. Also, there are discontinuities between the fields of views of adjacent cameras. In such a system, humans need to be tracked across discontinuous fields of views of multiple cameras. Our proposed model addresses this issue using the concepts of gait pattern, gait model, and motion signature. Each human's gait pattern is constructed and stored in a database. This gait pattern spans a tensor space that consists of three dimensions: person, image feature, and spatio-temporal data. A human's gait model can be constructed from the gait pattern using the "tensor decomposition based approach" described in this paper. When human(s) appears in one of the camera's field of a view (which is often discontinuous from the other camera's field of views), the human's motion signature is calculated and compared to each person in the database's gait model. The person with the gait model that is most similar to the motion signature is identified as same person. After the person is identified, the person is tracked within the field of view of the camera using the mean-shift algorithm based on color parameters. We conducted two experiments; the first experiment was identifying and tracking humans in a single video sequence, and experiments, the percentage of subjects that were correctly identified and tracked was better than that of two currently widely-used methods, PCA and nearest-neighbor. In the second experiment was the same as the first experiment but consisted of multiple-cameras with discontinuous views. The second experiment (human tracking across discontinuous images), shows the potential validity of the proposed method in a typical surveillance system.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • A Method for Detecting Multiple Independently Moving Objects from the Sequences Acquired by Active Stereo Cameras and Estimating the Cameras' Egomotion

    Yingdi Xie, Jun Ohya

    Journal of the Institute of Image Electronics Engineers of Japan   39 ( 2 ) 163 - 174  2010  [Refereed]

     View Summary

    This paper proposes a modified RANSAC based method that can detect multiple independently moving objects from the video sequences acquired by active stereo cameras, and can estimate the egomotion of the stereo cameras. We first compute 3D optical flows from consecutive frames, where dense 3D optical flows, which are needed for the subsequent egomotion estimation based on the modified RANSAC, are obtained by the process proposed in this paper. From the dense flows, three 3D optical flows are randomly selected for computing an egomotion (model). The validity of that egomotion model is checked using the weight obtained by calculating a consistency between that model and each of the dense flows. If it turns out that the estimated egomotion model is valid, the label that identifies this model is assigned to the inliers, which are consistent with this model, and then the labeled inliers are deleted. The random choice of three flows and its subsequent processes are repeated till one of the termination con itions is met. Using the inliers present in the static background, the real egomotion is computed. Experiments using synthesized and real stereo sequences demonstrate the effectiveness of the proposed method. © 2010, The Institute of Image Electronics Engineers of Japan. All rights reserved.

    DOI

    Scopus

    5
    Citation
    (Scopus)
  • An improved Kernel-based fuzzy C-means algorithm with spatial information for brain MR image segmentation

    Rong Xu, Jun Ohya

    International Conference Image and Vision Computing New Zealand   173   1 - 7  2010  [Refereed]

     View Summary

    In this paper, we propose an improved Kernel-based Fuzzy C-means Algorithm (iKFCM) with spatial information to reduce the effect of noise for brain MR image segmentation. We use k-nearest neighbour model and a neighbourhood controlling factor by estimating image contextual constraints to optimize the objective function of conventional KFCM method. Conventional KFCM algorithms classify each pixel in image only by its own gray value, but the proposed method classifies by the gray values of its neighbourhood system. For this reason, the proposed iKFCM has a strong robustness for image noise in image segmentation. In experiments, some synthetic grayscale images and simulated brain MR images are used to assess the performance of iKFCM in comparison with other fuzzy clustering methods. The experimental results show that the proposed iKFCM method achieves a better segmentation performance than other fuzzy clustering methods. © 2010 IEEE.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Feasibility of using machine vision based cash register system in a University Cafeteria

    Khanh N. Do, Jun Ohya, Davar Pishva, G. G D Nishantha

    Proceedings - 2010 3rd International Congress on Image and Signal Processing, CISP 2010   6   2554 - 2558  2010  [Refereed]

     View Summary

    This paper explores the feasibility of using machined vision based cash register system in a university cafeteria and automatically determine price of the food items that are contained on a cafeteria food tray. The idea is to take an image of the food tray, identify all the food items on the tray, and display the total food cost. A fully automated version of the system is envisioned to use a standard biometric technology for customer identification and for automatic debit process of the food cost from the customer's pre-registered account. The analysis method uses shape and size analysis to identify food items that are served on distinct type dishes and color analysis for those items that are served in similar type dishes. The paper explains prototype of a developed system and experimental results for 10 types of meals served at Ritsumeikan Asia Pacific University Cafeteria to justify the validity of this approach. ©2010 IEEE.

    DOI

    Scopus

  • Application of extended producer responsibility in Chinese WEEE management

    Yiping Zhao, Jun Ohya

    ICCCE 2010 - 2010 International Conference on Chemistry and Chemical Engineering, Proceedings     272 - 275  2010  [Refereed]

     View Summary

    In this paper, the administration-driven WEEE management system in Europe and the market-driven end-of-life automobile management system in the US are studied first in this paper. Successful experiences and deficiencies of each system are investigated and compared. Then the potentials for integrating these two systems under each single-instrument-dominant scenario are identified followed by the discussion on key factors for soundly building an integrated system under various scenarios in general. Based on these achievements, the WEEE management system in China is taken as an example to demonstrate how to practically set up an integrated system in such a specific conditions, so that the application of EPR principle could be better promoted in a more economic efficient and ecological effective manner. © 2010 IEEE.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • STUDY ON HUMAN GESTURE RECOGNITION FROM MOVING CAMERA IMAGES

    Dan Luo, Jun Ohya

    2010 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO (ICME 2010)     274 - 279  2010  [Refereed]

     View Summary

    We develop a framework based approach to extract and recognize hand gestures from the video sequence acquired by a dynamic camera, which could be a useful interface between humans and mobile robots. We use Human-Following Local Coordinate (HFLC) System, a very simple and stable method for extracting hand motion trajectories, which is obtained from the located human face and body part. Hand trajectory motion models (HTMM) are constructed by HFLC and hand blob changing factor. In this paper, we apply a principal component analysis (PCA) based approach to improve the recognition accuracy. For further improvement, temporal changes in the observed hand area changing factor are utilized as new image features to be stored in the database after being analyzed by PCA. Each HTMM in the database is classified into gesture categories, or temporal changes in hand blob changes. We demonstrate the effectiveness of the proposed method by conducting experiments on 51 kinds of sign language based Japanese and American Sign Language gestures obtained from 7 people. Our experimental recognition results show better performance is obtained by PCA based approach than the Condensation algorithm based method.

    DOI

    Scopus

    4
    Citation
    (Scopus)
  • Visual Support System Considering Privacy Protection by Using the Line Drawing Video Image

    Shun-Ichi Yonemurat, Ken Tsutsuguchit, Satoshi Shimadat, Akira Suzukit, Yukio Tokunaga, Jun Ohya

    Journal of the Institute of Image Electronics Engineers of Japan   39 ( 4 ) 481 - 489  2010  [Refereed]

     View Summary

    This paper proposes a system that automatically deforms the original video images into line drawing expression transmitted in both directions so as to achieve two goals: easing the concerns of users privacy and ensuring good enough support by visual channel. We examine the effect of line drawing video for support system on the effectiveness and impression of privacy on a collaboration between a novice user and an operator. As a result, it became clear that concern of the user on privacy or security are greatly reduced by using a line drawing image, and there was no significant difference on efficiency with an original image and a line drawing image. © 2010, The Institute of Image Electronics Engineers of Japan. All rights reserved.

    DOI

    Scopus

  • •” Media: Text Based CMC Tool Which Touch off Informal Communication

    Shun-Ichi Yonemura, Satoshi Shimada, Chen Li Jen, Jun Ohya, Yukio Tokunaga

    Journal of the Institute of Image Electronics Engineers of Japan   39 ( 4 ) 442 - 450  2010  [Refereed]

     View Summary

    This paper describes about the media which touch off informal text communication by black spot stimulus. A black spot stimulus is shown as a background image of the text field in advance of text input. The purpose of black spot stimulus presentation is to touch off conceiving topic based on the cognitive process model of creative thinking. The prototype system based on this media concept was developed, and the communication experiment was conducted. It turned out as a result of the experiment that these media touched off the idea about making topic strongly in the text communication. Moreover, it was implied that the topic touched off deepens the mutual understanding between speakers. © 2010, The Institute of Image Electronics Engineers of Japan. All rights reserved.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • 中国麦積山石窟のデジタルアーカイブとVRコンテンツ

    阿部信明, 河合隆史, 大谷 淳, 査 紅彬, 安藤 真

    日本バーチャルリアリティ学会論文誌   14 ( 3 ) 275 - 282  2009.09  [Refereed]

     View Summary

    Recently, activities about the 3D digital archive of cultural heritages become very popular. The development of this technology aims at satisfying the competing ideals of preserving cultural heritage while at the same time making it available to the public. Rich expression by the stereoscopic 3D is suitable for the viewing of 3D data saved culture heritages. In this paper, we report on the work of archiving heritage of Chinese Maijishan Grotto and the developing of stereoscopic VR content for it. In our content, viewers can experience the precious cultural heritage which can't be usually watched. Representation of cultural heritage can be more lifelike by presenting stereoscopic images and interactions. In addition, the results of the subjective evaluation confirmed VR content's reality and 3D effects.

    DOI CiNii

  • Elliptical Shaped Object Recognition via a Modified RANSAC with Edge Orientation Curve’s Segmentation-Merge

    Yingdi Xie, Jun Ohya

    Proc. Visualization, Imaging, and Image Processing   652-028   33 - 40  2009.07  [Refereed]

  • Company-Level Adaptive Management Model under Extended Producer Responsibility

    Yiping Zhao, Yingdi Xie, Jun Ohya

    Proceeding of The 24th International Conference on Solid Waste Technology and Management   24   804 - 813  2009.03  [Refereed]

  • Efficient Detection of Ellipses from an Image by a Guided Modified RANSAC

    Yingdi Xie, Jun Ohya

    IMAGE PROCESSING: ALGORITHMS AND SYSTEMS VII   7245   72450W  2009  [Refereed]

     View Summary

    In this paper, we propose a novel ellipse detection method which is based on a modified RANSAC, with automatic sampling guidance from the edge orientation difference curve. Hough Transform family is one of the most popular and methods for shape detection, but the Standard Hough Transform loses its computation efficiency if the dimension of the parameter space gets high. Randomized Hough Transform, an improved version of Standard Hough Transform has difficulty in detecting shapes from complicated, cluttered scenes because of its random sampling process. As a pre-process for random selection of five pixels to be used to build the ellipse's equation, we propose a two-step algorithm: (1) region segmentation and contour detection by mean shift algorithm (2) contour splitting based on the edge orientation difference curve obtained from the contour of each region. In each contour segment obtained by step (2), 5 pixels are randomly selected and the modified RANSAC is applied to the 5 pixels so that an accurate ellipse model is obtained. Experimental result show that the proposed method can achieve high accuracies and low computation cost in detecting multiple ellipses from an image.

    DOI

    Scopus

    6
    Citation
    (Scopus)
  • Comparative Study of Methods for Recognition an Unknown Person's Action from a Video Sequence

    Takayuki Hori, Jun Ohya, Jun Kurumisawa

    IMAGE PROCESSING: ALGORITHMS AND SYSTEMS VII   7245   72450V  2009  [Refereed]

     View Summary

    This paper proposes a Tensor Decomposition Based method that can recognize an unknown person's action from a video sequence, where the unknown person is not included in the database (tensor) used for the recognition. The tensor consists of persons, actions and time-series image features. For the observed unknown person's action, one of the actions stored in the tensor is assumed. Using the motion signature obtained from the assumption, the unknown person's actions are synthesized. The actions of one of the persons in the tensor are replaced by the synthesized actions. Then, the core tensor for the replaced tensor is computed. This process is repeated for the actions and persons. For each iteration, the difference between the replaced and original core tensors is computed. The assumption that gives the minimal difference is the action recognition result. For the time-series image features to be stored in the tensor and to be extracted from the observed video sequence, the human body silhouette's contour shape based feature is used. To show the validity of our proposed method, our proposed method is experimentally compared with Nearest Neighbor rule and Principal Component analysis based method. Experiments using 33 persons' seven kinds of action show that our proposed method achieves better recognition accuracies for the seven actions than the other methods.

    DOI

  • Hand-gesture extraction and recognition from the video sequence acquired by a dynamic camera using condensation algorithm

    Luo Dan, Jun Ohya

    Proceedings of SPIE - The International Society for Optical Engineering   7252   72520S  2009  [Refereed]

     View Summary

    To achieve environments in which humans and mobile robots co-exist, technologies for recognizing hand gestures from the video sequence acquired by a dynamic camera could be useful for human-to-robot interface systems. Most of conventional hand gesture technologies deal with only still camera images. This paper proposes a very simple and stable method for extracting hand motion trajectories based on the Human-Following Local Coordinate System (HFLC System), which is obtained from the located human face and both hands. Then, we apply Condensation Algorithm to the extracted hand trajectories so that the hand motion is recognized. We demonstrate the effectiveness of the proposed method by conducting experiments on 35 kinds of sign language based hand gestures. © 2009 SPIE-IS&amp
    T.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Prompter "center dot" Based Creating Thinking Support Communication System That Allows Hand-Drawing

    Li Jen Chen, Jun Ohya, Shunichi Yonemura, Sven Forstmann, Yukio Tokunaga

    HUMAN-COMPUTER INTERACTION, PT II   5611   783 - +  2009  [Refereed]

     View Summary

    Research into creative thinking-support tools and communication is commonly focused on how to develop and share ideas between participants or with others. In this paper, we proposes a creative thinking support method that utilizes randomly generated visual prompter (black circle) image patterns (VP-patterns) and free hand-drawing and writing functions. Concepts and ideas of the research have been explained together with the development of the systems (CSP1 and CSP2). Experiments have been Conducted in order to evaluate the potentials and effectiveness of the system. From the results, a tendency towards inspiring creative ideas by participants has been observed.

    DOI

    Scopus

  • 3D Sensation Enhancement of 3D Digitized Models for 2D Displays Using Local Gray-level Contrast and Range Information

    Xiao Ding, Jun Ohya

    2008KJPR(Third Korea-Japan Joint Workshop on Pattern Recognition)     55 - 56  2008.11  [Refereed]

  • Contrast Enhancement of Color Images based on PCA of Luminance and Saturation

    Xiao Ding, Jun Ohya

    2008KJPR(Third Korea-Japan Joint Workshop on Pattern Recognition)     9 - 10  2008.11  [Refereed]

  • Development and Evaluation of the System for Educational Method Using Students’ Understanding Degree

    Takaaki Ohkawauchi, Jun Ohya, Yoshikazu Tateno

    ICoME 2008“Enhancing ICT in Education for Global Society”     266 - 273  2008.08  [Refereed]

  • Body-part Segmentation for a Real Human Wearing Multiple-Colored Suit by a Mean Shift Based Algorithm

    Huimin Guo, Jun Okazaki, Dong-Wan Kang, Jun Ohya

    2007KJPR (Second Korea-Japan Joint Workshop)    2007.10  [Refereed]

  • Accurate, Real-time 3D Reconstruction of Human Postures from Orthogonal Images Based on Neural Networks

    Dong-Wan Kang, Zhongze Chen, Jun Ohya

    2007KJPR (Second Korea-Japan Joint Workshop)     111 - 116  2007.10  [Refereed]

  • ACCESSIBLE INFORMATION PRESENTATION METHOD IN PUBLIC SPACE

    Shunichi Yonemura Shinichiro Eitoku Kenichiro Shimokura, Jun Ohya

    Proceedings of Workshop on Tactic and Haptic Interaction     70 - 75  2007.05  [Refereed]

  • MULTIMODAL COMMUNICATION ON VISUAL SUPPORT SYSTEM

    Shunichi Yonemura, Tohru Yoshida, Yukio Tokunaga, Jun Ohya

    Proceedings of Workshop on Tactic and Haptic Interaction     64 - 69  2007.05  [Refereed]

  • A STUDY OF A COMPUTER MEDIATED COMMUNICATION VIA THE “●” PROMPT SYSTEM - Introducting the affects on the stimulation of thought processes and the inspiration of creative ideas -,

    Li Jen Chen, Jun Ohya, Shunichi Yonemura, Nobuyuki, Harada,Yukio Tokunaga

    Proceedings of Workshop on Tactic and Haptic Interaction     50 - 55  2007.05  [Refereed]

  • 逆S字型変換による濃淡画像のコントラスト強調手法

    小林裕一, 加藤俊一, 大谷 淳

    電子情報通信学会論文誌 D   Vol. J90-D ( No. 5 ) 1263 - 1274  2007.05  [Refereed]

  • 境界要素法に基づく3次元多関節弾性物体の効率的かつ頑健な挙動シミュレーション

    渡辺隆史, 大谷 淳

    日本バーチャルリアリティ学会論文誌   12 ( 1 ) 85 - 94  2007.03  [Refereed]

     View Summary

    This paper proposes a new efficient, robust, BEM (Boundary Element Method) based method for simulating articulated 3D elastic objects' behaviors. The articulated elastic object consists of bones, elastic object primitives and joining surfaces. After transforming the bones by inverse kinematics, the elastic object primitives are deformed basically by the BEM. To reduce the computation cost and memory usage and to achieve robust deformations, this paper improves the BEM based module of the conventional method: more specifically, a new method for computing the displacements of the vertices in the joining surfaces. Experimental results show that the proposed model can be simulated 10 times as fast as the conventional model or faster, and that real time performances such as 60fps can be achieved even for models with 20,000 vertices. High quality simulations of different deformations including expansions and contractions are achieved using character type models.

    DOI CiNii

  • Efficient multiple independent motion segmentation from an active platform by utilizing modified RANSAC

    Yingdi Xie, Jun Ohya

    PROCEEDINGS OF THE IASTED INTERNATIONAL CONFERENCE ON GRAPHICS AND VISUALIZATION IN ENGINEERING     31 - +  2007  [Refereed]

     View Summary

    hi this paper, an efficient approach to segmentation of different independent motion areas from a moving platform is described. This approach is implemented on a stereo vision system, depth information could be computed by matching feature points between stereo images. For two consecutive frames, ego-motion is estimated from the optical flows, including depth information belonging to the background, which has a larger space distribution comparing to those of independent moving objects. In order to distinguish different motion areas, we proposed a modified version of RANSAC mechanism, which could handle the problem of multiple model extraction in a noisy environment.

  • Contrast enhancement of color images based on wavelet transform and human visual system

    Ding Xiao, Jun Ohya

    PROCEEDINGS OF THE IASTED INTERNATIONAL CONFERENCE ON GRAPHICS AND VISUALIZATION IN ENGINEERING     58 - +  2007  [Refereed]

     View Summary

    This paper proposes a new method for enhancing the contrast of color images based on Wavelet Transform and human visual system. The RGB (red, green, and blue) values of each pixel in a color image are converted to HSV (hue, saturation and value) values. To the V (luminance value) components of the color image, Wavelet Transform is applied so that the V components are decomposed into the approximate components and detail components. The obtained coefficients of the approximate components are converted by a grey-level contrast enhancement technique based on human visual system. Then, inverse Wavelet transform is performed for the converted coefficients so that the enhanced V values are obtained. The S components are enhanced by histogram equalization. The H components are not changed, because changes in the H components could degrade the color balance between the HSV components. The enhanced S and V together with H are converted back to RGB values. The effectiveness of the proposed method is demonstrated experimentally.

  • Efficient, robust, bem (boundary element method) based method for simulating articulated 3d elastic object models

    Takafumi Watanabe, Jun Ohya

    PROCEEDINGS OF THE IASTED INTERNATIONAL CONFERENCE ON GRAPHICS AND VISUALIZATION IN ENGINEERING     7 - +  2007  [Refereed]

     View Summary

    This paper proposes an efficient and effective computation method for simulating articulated 3D elastic objects. The articulated elastic object consists of bones, elastic object primitives and joining surfaces. After transforming the bones by inverse kinematics, the elastic object primitives are deformed basically by the BEM (Boundary Element Method). To reduce the computation cost and memory usage and to achieve robust deformations, this paper improves the BEM based module of the conventional method: more specifically, a new method for computing the displacements of the vertices in the joining surfaces. Experimental results show that the proposed model can be simulated 10 times as fast as the conventional model or faster, and that real-time performances such as 60fps can be achieved even for models with 20,000 vertices. High quality simulations of different deformations including expansions and contractions are achieved using character type models.

  • Deformation Styles for Spline-based Skeletal Animation

    Sven Forstmann, Jun Ohya, Artus Krohn-Grimberghe, Ryan McDougall

    SYMPOSIUM ON COMPUTER ANIMATION 2007: ACM SIGGRAPH/ EUROGRAPHICS SYMPOSIUM PROCEEDINGS     141 - 150  2007  [Refereed]

     View Summary

    We present a novel skinned skeletal animation system based on spline-aligned deformations for providing high quality and fully designable deformations in real-time. Our ambition is to allow artists the easy creation of abstract, pose-dependent deformation behaviors that might directly be assigned to a large variety of target objects simultaneously To achieve this goal, we introduce the usage of deformation styles and demonstrate their applicability by our animation system. We therefore enhance spline-skinned skeletal animation with two sweep-based free-form-deformation (FFD) variants. The two FFD variants are pose-dependent, driven by three textures and three curves, which can be designed by the artist. As the three textures are similar to height-maps, their creation is very intuitive. Once designed, the deformation styles can be directly applied to any number of targets for imitating material behaviors of cloth, metal or even muscles. Our GPU based implementation shows promising results for real-time usage, as about 30 Million vertices per second can be animated. The basic spline-skinning even reaches more than twice the speed and gets close to the performance of skeletal subspace deformation (SSD). Furthermore, our method can easily be combined along with other existing deformation techniques as pose space deformation or SSD.

  • Communication analysis of visual support system that uses line drawing expression

    Shunichi Yonemura, Tohru Yoshida, Yukio Tokunaga, Jun Ohya

    HUMAN INTERFACE AND THE MANAGEMENT OF INFORMATION: METHODS, TECHNIQUES AND TOOLS IN INFORMATION DESIGN, PT 1, PROCEEDINGS   4557   359 - 365  2007  [Refereed]

     View Summary

    This paper proposes a system that automatically deforms the images transmitted in both directions so as to achieve two goals: easing the concerns of users and ensuring well enough support. We examine here the effect of visual information quality on the effectiveness and impression of collaboration between a novice user and an operator.

  • A study of a computer mediated communication via the "circle" prompt system - Analysis of the affects on the stimulation of thought processes and the inspiration of creative ideas

    Li Jen Chen, Nobuyuki Harada, Shunichi Yonemura, Jun Ohya, Yukio Tokunaga

    2007 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOLS 1-5     1119 - +  2007  [Refereed]

     View Summary

    Research into thinking-support tools is commonly focused on how to develop and share ideas between participants or with others. In this paper, we propose and develop a communication system that stimulates the thought processes and inspires the creative ideas of participants by using a visual "circle" prompt within the framework of a communication pallet. Experiments have been conducted into methods of stimulating the thought process and inspiring ideas during conversation and the results have been analyzed. From the results, a tendency towards inspiring creative ideas by participants has been observed.

  • A study of a computer mediated communication via the "circle" prompt system - Analysis of the affects on the stimulation of thought processes and the inspiration of creative ideas

    Li Jen Chen, Nobuyuki Harada, Shunichi Yonemura, Jun Ohya, Yukio Tokunaga

    2007 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOLS 1-5     1119 - +  2007  [Refereed]

     View Summary

    Research into thinking-support tools is commonly focused on how to develop and share ideas between participants or with others. In this paper, we propose and develop a communication system that stimulates the thought processes and inspires the creative ideas of participants by using a visual "circle" prompt within the framework of a communication pallet. Experiments have been conducted into methods of stimulating the thought process and inspiring ideas during conversation and the results have been analyzed. From the results, a tendency towards inspiring creative ideas by participants has been observed.

  • 伸縮を含む対話的変形操作が可能な多関節三次元弾性物体モデル

    渡辺隆史, 大谷 淳, 楜沢 順, 徳永幸生

    電子情報通信学会論文誌D   J89-D ( 10 ) 2145 - 2154  2006.10  [Refereed]

  • やわらかさ・かたさの判別に係る視覚的要因に関する一考察

    小林裕一, 大谷 淳

    GITS/GITI紀要、早稲田大学大学院国際情報通信研究科/国際情報通信研究センター     178 - 187  2006.07  [Refereed]

  • VISUAL BLENDING USING 2D MORPHING

    S.K. Semwal, J. Ohya, I. Semwal, B Bikker

    Proceedings of the 24th IASTED International Multi-Conference Signal Processing, Pattern Recognition, and Applications     316 - 321  2006.02  [Refereed]

  • Fast Skeletal Animation by skinned Arc-Spline based Deformation

    Sven Forstmann, Jun Ohya

    EUROGRAPHICS 2006: SHORT PAPERS     1 - 4  2006  [Refereed]

     View Summary

    Presented is a novel skeletal animation system for providing high quality geometric deformations in real-time. Each bone of the skeleton is therefore represented by a spline, rather than using conventional matrix rotation. In our approach, each vertex of the animated character can be influenced by a maximum of three spline-curves, which is sufficient for skinned animation. One spline is parameterized by three control points and a possible twist. As opposed to conventional Bezier curves does our arc-spline rely on trigonometric functions for providing better curvatures. The optimized implementation using the OpenGL-shading language shows very promising results for real-time character animation, as even about 1 Million vertices were able to be transformed at interactive 43 frames per second on a GeForce 7800 GTX graphics card.

  • EM-in-M: Analyze and synthesize emotion in motion

    Yuichi Kobayashi, Jun Ohya

    ADVANCES IN MACHINE VISION, IMAGE PROCESSING, AND PATTERN ANALYSIS   4153   135 - 143  2006  [Refereed]

     View Summary

    We have been researching the relationship between human motion and emotion. In this paper, our purpose is to extract motion features specific to each emotion. We propose a new approach for motion data analysis, which applies the higher order Singular Value Decomposition(HOSVD) direct to the motion data and the wavelet analysis to the synthesized data with SVD. The HOSVD models the mapping between persons and emotions. The model can synthesize a complete data acting with each emotion for a given new person. The wavelet analysis extracts each motion feature from the synthesized data for each emotion. Some experimental results using motion capture data for "gait" action and 6 emotions - "angry, joy, sad and so on" show that our method can synthesize novel gait motions for a person by using the extracted motion elements and can extract some features specific to each emotion.

  • 2段階境界要素法を用いる三次元弾性物体の変形と移動の実時間アニメーション法

    渡辺隆史, 大谷 淳, 楜沢 順, 徳永幸生

    電子情報通信学会論文誌 D-Ⅱ   J88-D-Ⅱ ( 9 ) 1876 - 1888  2005.09  [Refereed]

  • Visualization of large iso-surfaces based on nested clip-boxes

    Sven Forstmann, Jun Ohya

    ACM SIGGRAPH 2005 Posters, SIGGRAPH 2005   Conference Select CD-ROM Disc   126  2005.07  [Refereed]

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • 顔面筋動作に基づく形状変形とテクスチャの張替えによる三次元人物顔画像の再構築

    佐藤一, 大谷 淳, 寺島信義

    電子情報通信学会論文誌 D-Ⅱ   J88-D-Ⅱ ( 7 ) 1126 - 1142  2005.07  [Refereed]

  • Visual-dimension interact system (VIS)

    A Onda, T Oku, E Yu, Y Lee, Choh, I, PY Chiu, J Ohya

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E88D ( 5 ) 947 - 953  2005.05  [Refereed]

     View Summary

    In this paper we describe a mixed reality-supported interactive viewing enhancement museum display system: Visual-dimension Interact System (VIS). With a transparent interactive inter-face, the museum visitor is able to see, manipulate, and interact with the physical exhibit and its virtual information, which are overlapped on one other. Furthermore, this system provides the possibility for visitor to experience the creation process in an environment as close as possible to the real process. This has the function of assisting the viewer in understanding the exhibit and most importantly, gaining a so-to-speak hands-on experience of the creation process itself leading to a deeper understanding of it.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Analysis of expressing audiences in a cyber-theater

    K DongWan, K Huang, J Ohya

    2005 IEEE International Conference on Multimedia and Expo (ICME), Vols 1 and 2   CD-ROM Proceedings, 4 pages   727 - 730  2005  [Refereed]

     View Summary

    This paper studies how audiences should be expressed in a Cyber-theater, in which remotely located persons can direct plays as directors, perform as performers and/or see the performances as audiences through a networked virtual environment. It is noted that the audience effect has been widely acknowledged in the real-world theater: that is, the audience reaction has a significant effect on the acting of player and performance of the play itself. However, only a few works relevant to audiences in the Cyber Theater can be seen. This paper studies whether the audience effect exists also in the Cyber-theater. By constructing a system in which two actors are displayed a remotely located audience's avatar in which the audience can display his/her emotional actions, we clarified that interactions between the actors and audience are effective.

    DOI

    Scopus

  • A study of synthesizing new human motions from sampled motions using tensor decomposition

    R Kalanov, J Cho, J Ohya

    2005 IEEE International Conference on Multimedia and Expo (ICME), Vols 1 and 2   CD-ROM Proceedings, 4 pages   1327 - 1330  2005  [Refereed]

     View Summary

    This paper applies an algorithm, based on Tensor Decomposition, to a new synthesis application: by using sampled motions of people of different ages under different emotional states, new motions for other people are synthesized Human motion is the composite consequence of multiple elements, including the action performed and a motion signature that captures the distinctive pattern of movement of a particular individual. By performing decomposition, based on N-mode SVD (singular value decomposition), the algorithm analyzes motion data spanning multiple subjects performing different actions to extract these motion elements. The analysis yields a generative motion model that can synthesize new motions in the distinctive styles of these individuals. The effectiveness of applying the tensor decomposition approach to our purpose was confirmed by synthesizing novel walking motions for a person by using the extracted signature.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • “Network Transmission of Human Face Data and 3D Reconstruction Considering Image Data Reduction”

    Hajime Sato, Jun Ohya, Nobuyoshi Terashima

    Journal of the Institute of Image Electronics Engineers of Japan   34 ( 5 ) 539 - 547  2005  [Refereed]

     View Summary

    A method for enabling efficient facial data transmission over the communication network and reconstructing realistic 3D face images is proposed. A conventional method which combines the advantages of the model-based and image-based approaches to reconstruct 3D facial images with high efficiency and precision on a standalone environment is extensively used. In our approach, the face image recognition part and the 3D image reconstruction part, which were conventionally carried out on the same PC, are separated and implemented on two network-connected PCs. Also, by using JPEG compression, the size of image data transmitted over the network is reduced. Our method was implemented on network-connected PCs, and its effectiveness was confirmed. Subjective tests showed that image data could be reduced to around 6% of its original size. © 2005, The Institute of Image Electronics Engineers of Japan. All rights reserved.

    DOI

    Scopus

  • Editorial - Special section on multimodal interfaces and applications

    J Ohya, R Reilly, CS Li

    IEEE TRANSACTIONS ON MULTIMEDIA   6 ( 3 ) 397 - 397  2004.06

    DOI

    Scopus

  • 強化現実技術を応用した立体ディスプレイシステムの開発

    盛川浩志, 川口真実, 河合隆史, 大谷淳

    日本人間工学会誌   第40巻特別号   372 - 373  2004.06

    DOI CiNii

  • Human motion tracking using nonsynchronous multiple observations

    H Mori, A Utsumi, J Ohya, M Yachida, R Nakatsu

    ELECTRONICS AND COMMUNICATIONS IN JAPAN PART III-FUNDAMENTAL ELECTRONIC SCIENCE   87 ( 12 ) 84 - 93  2004  [Refereed]

     View Summary

    This paper proposes a method of tracking a human object by using nonsynchronous multiple-viewpoint images. The proposed method tracks human forms efficiently by using a Kalman filter to integrate observed information which is obtained nonsynchronously from multiple viewpoints. The experimental system is composed of multiple observation nodes, which operate nonsynchronously to process the multiviewpoint images, a tracking node, which tracks the human figure, and the discovering node, which finds the human figure. The image features are matched to the tracking model in the observation node based on the prediction of the observed value which is sent from the tracking node. The image features matched to the model are sent to the tracking node, and the tracking model is updated. The image features which are not matched are sent to the discovering node to find a new human figure. With the proposed approach, it is possible to construct a large-scale tracking system while reducing the deterioration of processing efficiency and the redundancy among observations that occur in synchronous systems. The effectiveness of the proposed method is demonstrated by an experiment using real images. (C) 2004 Wiley Periodicals, Inc.

    DOI

    Scopus

  • Visual-dimension interact system (VIS) - Exhibiting creative process for museum visitor experience

    A Onda, T Oku, PY Chiu, E Yu, M Yokoi, Choh, I, J Ohya

    2004 INTERNATIONAL CONFERENCE ON CYBERWORLDS, PROCEEDINGS     278 - 285  2004  [Refereed]

     View Summary

    In this paper we describe a mixed reality supported interactive viewing enhancement museum display system. With a transparent interactive interface, the museum visitor is able to see, manipulate, and interact with the physical exhibit and its virtual information, which are overlapped on one other. Furthermore, this system provides the possibility for the visitor to experience the creation process in an environment as close as possible to the real process. This has the function of assisting the viewer in understanding the exhibit and most importantly, gaining a so-to-speak hands-on experience of the creation process itself leading to a deeper understanding of it.

  • Softness perceptive texture method

    Y Kobayashi, M Suzuki, J Ohya

    HUMAN VISION AND ELECTRONIC IMAGING IX   5292   322 - 329  2004  [Refereed]

     View Summary

    We have been studying about people's visual impression and image features for texture images in order to clarify the human subjective interpretation mechanism for images[1]. In corresponding image features of human impressions for the images, we found that the impressions for material were bottle-necked. We have studied a new analysis method which gives the impression for material from texture images. Especially, we mainly focused on the properties of visual targets which people can feel tactile sense. In this paper, we propose a new texture analysis method which is based on frequency analysis with 3D texture which is designed for photorealistic rendering. We found that our new method can estimate not only the surface roughness but also the surface softness.

  • Development of a stereoscopic 3D display system to observe restored heritage

    H Morikawa, M Kawaguchi, T Kawai, J Ohya

    STEREOSCOPIC DISPLAYS AND VIRTUAL REALITY SYSTEMS XI   5291   415 - 422  2004  [Refereed]

     View Summary

    The authors have developed a binocular-type display system that allows digital archives of cultural assets to be viewed in their actual environment. The system is designed for installation in locations where such cultural assets were originally present. The viewer sees buildings and other heritage items as they existed historically by looking through the binoculars. Images of the cultural assets are reproduced by stereoscopic 3D CG in cyberspace, and the images are superimposed on actual images in real-time. This system consists of stereoscopic CCD cameras that capture a stereo view of the landscape and LCDs for presentation to the viewer. Virtual cameras. used to render CG images from digital archives. move in synchrony with the actual cameras, so the relative position of the CG images and the landscape on which they are superimposed is always fixed. The system has manual controls for digital zoom. Furthermore, the transparency of the CG images can be altered by the viewer. As a case study for the effectiveness of this system, the authors chose the Heijyoukyou ruins in Nara, Japan. The authors evaluate the sense of immersion, stereoscopic effect, and usability of the system.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Real-time stereo by using dynamic programming

    Sven Forstmann, Yutaka Kanou, Jun Ohya, Sven Thuering, Alfred Schmitt

    IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops   2004- ( January )  2004  [Refereed]

     View Summary

    A Dynamic Programming (DP) based algorithm that achieves real-time, high quality stereo-matching is presented. A special coarse to fine approach as well as the MMX based assembler implementation mainly contribute to the reached computation speed. A novel vertical smoothing approach by re-integrating paths inside the DP matching allows the avoidance of noisy horizontal strokes, so that high quality stereo-matching is achieved. The current implementation capable of running at about 30 FPS on an 2.2GHz PC, which is sufficient for the utilization in real-time applications.

    DOI

    Scopus

    113
    Citation
    (Scopus)
  • Exploiting the cognitive synergy between different media modalities in multimodal information retrieval

    ZF Zhang, RF Zhang, J Ohya

    2004 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXP (ICME), VOLS 1-3   CD-ROM proceedings (4 pages)   2227 - 2230  2004  [Refereed]

     View Summary

    This is a position paper reporting an on-going collaboration project between SUNY Binghamton, USA, and Waseda University, Japan, on multimodal information retrieval through exploiting the cognitive synergy across the different modalities of the information to facilitate an effective retrieval. Specifically we focus on image retrieval in the applications where imagery data appear along with collateral text. It is noted that these applications are ubiquitous. We have proposed the Synergistic Indexing Scheme (SIS) to explicitly exploit the synergy between the information of imagery and text modalities. Since the synergy we have exploited between the information of imagery and text modalities is subjective and depends on specific cognitive context, we call this type of synergy as cognitive synergy. We have reported part of the empirical evaluation and are in the process to fully implement the SIS prototype for an extensive evaluation.

  • Cognitive bridge between haptic impressions and texture images for subjective image retrieval

    Y Kobayashi, J Ohya, ZF Zhang

    2004 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXP (ICME), VOLS 1-3   CD-ROM proceedings (4 pages)   2239 - 2242  2004  [Refereed]

     View Summary

    As a step towards subjective image retrieval, this paper reports an on-going collaboration project between Waseda University and SUNY Binghamton, on relating texture images to haptic impressions. To grasp the surface height variations, texture images are taken under different illuminations and viewing conditions. Our method applies a new frequency analysis method to the texture images. We evaluate the performances of our feature and other typical conventional features by checking whether texture images are correctly classified into "soft" or "hard" by the SVM (support vector machine) method, where the training data for the SVM are collected by subjective tests. Experimental results show that our texture feature can classify "soft" or "hard" better than the other features.

  • Computer vision based analysis of the botanical tree's dynamical behaviors for the reproduction in virtual space

    LC Lu, J Ohya

    2004 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXP (ICME), VOLS 1-3   CD-ROM proceedings (4 pages)   839 - 842  2004  [Refereed]

     View Summary

    This paper deals with a method that analyzes a botanical tree's behaviors in real space by a computer vision approach so as to reproduce the analyzed behaviors in virtual space. Instead of applying unstable local tracking to the tree in a video sequence, we estimate the direction and strength of the wind that shakes the tree by a learning based method that classifies the input video sequence into one of the stored winds with different directions and strengths. In the learning phase, sample video sequences are used for constructing the Eigenspace and Fisherspace, which is obtained from Fisher discriminant analysis. In the classification phase, the input video sequence is compared with each of the stored sample sequences so that the direction and strength of the wind are estimated. An interpolation method improves the estimation accuracy. Experimental results demonstrate the effectiveness of the proposed method.

  • Realistic 3D facial animation using parameter-based deformation and texture remapping

    H Sato, J Ohya, N Terashima

    SIXTH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE AND GESTURE RECOGNITION, PROCEEDINGS     735 - 742  2004  [Refereed]

     View Summary

    A method for constructing 3D human facial images which maintains the chararcteristics of video input image is proposed. First, a personalized 3D polygonal facial model is constructed from two orthogonal facial images. Video input frames are modified, and the vertices composing the 3D polygons are altered according to the contraction rule of facial muscles. Moreover, the modified input frame is texture mapped directly onto to the forementioned model to improve the quality of the reconstructed image. This method was implemented onto an ordinary PC, and an experiment using five persons' face data showed the effectiveness of the method both quantitatively and subjectively.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Estimating complicated and overlapped human body postures by wearing a multiple-colored suit using color information processing

    DW Kang, Y Onuma, J Ohya

    SIXTH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE AND GESTURE RECOGNITION, PROCEEDINGS     687 - 692  2004  [Refereed]

     View Summary

    This paper suggested an image processing method using a multiple-colored suit for estimating complicated postures of a human body. One of the major problems on the posture estimation when using the silhouette image analysis is the overlap of the body parts' silhouettes. In order to solve this problem, the human wears the multiple-colored suit so that each body part has a different color from the other body parts. Experimental results show the effectiveness of the proposed method. It has confirmed that the complicated overlapped postures can be analyzed through the contour of the human's silhouette and the color information from the body segments that are taken using three directional cameras which are located at the front, left, and right side of the subject.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Real-time Method for Animating Elastic Objects' Behaviors Including Collisions

    The 13th International Conference on Artificial Reality and Telexistence     270 - 277  2003.12  [Refereed]

  • The Study of Communicability of Verbal and Non-Verbal expressions in Haiku and Haiga

    Proceedings of The 4th Japan-Korea International Symposium on Kansei Engineering     313 - 314  2003.10  [Refereed]

  • Computer vision based recognition of interactions between human body and object

    Masumi Kobana, Jun Ohya

    IFIP Advances in Information and Communication Technology   112   173 - 180  2003  [Refereed]

     View Summary

    This paper proposes a computer vision based method that recognizes interactions between human body and object. In two successive frames in a video sequence, our function based contour analysis method checks whether the silhouettes of the human body and object overlap. If they do not overlap, our method judges that the human body and object does not interact. If the two silhouettes overlap, our method checks whether the motion vectors obtained from the areas for the human body and object coincide. If they coincide, our method judges that the human body and object are interacting. Some experimental results show the effectiveness of the proposed method. © 2003 by Springer Science+Business Media New York.

    DOI

  • Estimating postures of a human wearing a multiple-colored suit based on color information processing

    DW Kang, J Ohya

    2003 INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I, PROCEEDINGS   Ⅰ   261 - 264  2003  [Refereed]

     View Summary

    This paper proposes a non-contact type method for estimating human body postures. One of the major problems on the posture estimation using the silhouette image analysis is the overlapping of the body parts' silhouettes. In order to solve this problem, this paper proposes a method for estimating the posture of a human wearing a multiple-colored suit based on color information processing. By analyzing the contour of the human's silhouette, the method judges whether feature points are occluded by another body parts. If the occlusions occurs, color region segmentation is performed in order to know which region is frontal. The feature point in the frontal region is located in the skeleton of the region. Experimental results show the effectiveness of the proposed method.

    DOI

    Scopus

    8
    Citation
    (Scopus)
  • Efficient, realistic method for animating dynamic behaviors of 3D botanical trees

    H Kanda, J Ohya

    2003 INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL II, PROCEEDINGS   Ⅱ   89 - 92  2003  [Refereed]

     View Summary

    This paper proposes anew efficient method that can animate botanical trees in 3D realistically. In this paper, a 3D botanical tree model consists of a set of branch segments, to which leaf models are attached. To reduce the amount of computation, instead of calculating the motions of all the branch segments, only the representative segment in each branch is numerically analyzed. The numerical analysis is constrained to a 2D plane so that 3D numerical analysis need not be performed. Concerning the leaf model, a set of four leaves is systematically attached to each branch segment. Experimental results clarify the conditions for real-time, realistic animations of dynamic behaviors of trees.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Construction of realistic three dimensional facial images using video images

    H Sato, J Ohya, N Terashima

    SEVENTH INTERNATIONAL SYMPOSIUM ON SIGNAL PROCESSING AND ITS APPLICATIONS, VOL 2, PROCEEDINGS     227 - 230  2003  [Refereed]

     View Summary

    A method for constructing 3D human facial images which maintains the chararcteristics of video input image is proposed. First, a personalized 3D polygonal facial model is constructed from two orthogonal facial images. Video input frames are modified, and the movements of facial feature points are tracked and measured based on the the contraction rule of facial muscles. The vertices of the 3D polygons are altered accordingly, and the modified input frame is texture mapped directly onto to the forementioned model. This method was implemented using an ordinary CCD camera and computer, and an experiment using one person's data showed that realistic facial images could be reconstructed with high speed and precision.

    DOI

    Scopus

  • Material Analysis for Texture images based on Frequency Analysis

    KOBAYASHI Yuichi, OHYA Jun, TADENUMA Makoto

    ITE Technical Report   26   17 - 20  2002

     View Summary

    In order to correspond image features to human impressions for images, the impression for material is a bottle-neck. We have studied the new analysis method which gives the impression for material from a texture image. In this report, we propose a new material analysis method based on frequency analysis. Our method can classify texture images into the material which has similar surface structure.

    DOI CiNii

  • Analyzing Video Sequences of Multiple Humans - Tracking, Posture Estimation and Behavior Recognition.

    Jun Ohya, Akira Utsumi, Junji Yamato

      3   i-xxii, 1-138  2002  [Refereed]

    DOI

  • Face posture estimation using eigen analysis on an IBR (image based rendered) database

    Kuntal Sengupta, Philip Lee, Jun Ohya

    Pattern Recognition   35 ( 1 ) 103 - 117  2002.01  [Refereed]

     View Summary

    In this paper, we present a novel representation of the human face for estimating the orientation of the human head in a two dimensional intensity image. The method combines the use of the much familiar eigenvalue based dissimilarity measure with image based rendering. There are two main components of the algorith described here: The offline hierarchical image database generation and organization, and the online pose estimation stage. The synthetic images of the subject's face are automatically generated offline, for a large set of pose parameter values, using an affine coordinate based image reprojection technique. The resulting database is formally called as the IBR (or image based rendered) database. This is followed by the hierarchical organization of the database, which is driven by the eigenvalue based dissimilarity measure between any two synthetic image pair. This hierarchically organized database is a detailed, yet structured, representation of the subject's face. During the pose estimation of a subject in an image, the eigenvalue based measure is invoked again to search the synthetic (IBR) image closest to the real image. This approach provides a relatively easy first step to narrow down the search space for complex feature detection and tracking algorithms in potential applications like virtual reality and video-teleconferencing applications. © 2001 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • Construction of facial expressions using a muscle-based feature model

    YC Liu, H Sato, N Terashima, J Ohya

    IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I AND II, PROCEEDINGS   CD-Rom Proceedings 4 pages   469 - 472  2002  [Refereed]

     View Summary

    An efficient method for constructing facial images for use in telecommunication applications is proposed. This method uses a simple 3D feature model, which consists of polygons, which describe the shape of the face, and elastic linear springs, which simulate the natural movements of facial muscles. This method requires only two orthogonal facial images, and could easily be implemented on a relatively low-spec PC. Experimental results showed good results that various facial expressions could be synthesized and displayed from arbitrary directions.

  • Analysis of human behaviors by computer vision based approaches

    J Ohya

    IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I AND II, PROCEEDINGS   CD-Rom Proceedings 4 pages   913 - 916  2002  [Refereed]

     View Summary

    This paper describes the author's recent activities related to computer vision based methods for analyzing human behaviors: more specifically, posture estimation and recognizing interactions between a human body and object. For estimating postures in 3D from multiple camera images, the authors developed a heuristic based method and non-heuristic method. The heuristic based method heuristically analyzes the contour of a human silhouette so that significant points of a human body can be located in each image. The non-heuristic method utilizes a function for analyzing contours without using heuristic rules. Recognizing the interactions exploits the function based contour analysis and motion vector based analysis so that the system can judge whether the human body interacts with the object.

  • Spatial filtering using the active-space indexing method

    SK Semwal, J Ohya

    GRAPHICAL MODELS   63 ( 3 ) 135 - 150  2001.05  [Refereed]

     View Summary

    Traditionally most camera-based position estimation systems use only a few points to calibrate cameras. In this paper, we investigate a novel and alternate approach for 3D position estimation by using a larger number of points arranged in a 3D grid. We present an implementation of the active-space indexing mechanism which uses three cameras. Given the corresponding points in camera images, a precise estimation of the position can be obtained. The active-space indexing method can be also used as a spatial filter to eliminate the large number of possible corresponding pairs from consideration. This capability, unique only to the active-space indexing method, provides a tractable algorithm to the otherwise intractable situation. (C) 2001 Academic Press.

    DOI

    Scopus

  • User-Guided Composition Effects For Art-Based Rendering

    2001 ACM Symposium on Interactive 3D Graphics, ACM SIGGRAPH     99 - 102  2001.03  [Refereed]

  • 非同期多視点画像による人物追跡システムの構築

    電子情報通信学会論文誌D-II   Vol. J84-D-II ( No.1 ) 102 - 110  2001.01  [Refereed]

  • Computer vision based analysis of non-verbal information in HCI

    Jun Ohya

    Proceedings - IEEE International Conference on Multimedia and Expo     869 - 872  2001  [Refereed]

     View Summary

    This paper overviews our research activities on computer vision based non-verbal information analysis that can be applied to virtual communication environments and human computer interactions. In virtual communication environments, a user's facial expressions and body motions are estimated by computer vision approaches, and the estimated non-verbal information is reproduced in the user's avatar. For human computer interfaces, hand gestures are recognized as pre-defined commands by analyzing multiple camera images that observe the hand. In addition, facial expressions and body gestures are recognized from a time-sequential images by HMM (Hidden Markov Models).

    DOI

    Scopus

  • Remarks on a real-time, noncontact, nonwear, 3D human body posture estimation method

    Kazuhiko Takahashi, Tatsumi Sakaguchi, Jun Ohya

    Systems and Computers in Japan   31 ( 14 ) 1 - 10  2000.12  [Refereed]

     View Summary

    This paper proposes a new real-time method of estimating human postures in three dimensions from trinocular images. The proposed method extracts feature points of the human body by analyzing contours of human silhouettes. The feature points are extracted by using the subtraction images when self-occlusions occur in the silhouette images. Dynamic compensation is carried out with a Kalman filter so that all feature points are tracked. The 3D coordinates of the feature points are reconstructed by considering the geometrical relationship between the three cameras. Experimental results confirm both the feasibility and the effectiveness of the proposed method.

    DOI

    Scopus

    9
    Citation
    (Scopus)
  • Computer Vision Based Analysis of Non-verbal Information

    Jun Ohya

    Sixth International Conference on Control, Automation, Robotics and VIsion   CD-ROM Proceedings 6 pages  2000.12  [Refereed]

  • Virtual Environments for Human-to-Human Communication

    Jun Ohya

    IEEE International Conference on Multimedia and Expo   Tutorial 4B Notes  2000.07  [Refereed]

  • Bunshin: A Believable Avatar Surrogate for Both Scripted and On-the-Fly Pen-Based Control in a Presentation Environment

    Sketches & Applications, SIGGRAPH2000 Conference Abstracts and Applications     187  2000.07  [Refereed]

  • Remarks on Real-Time Human Body Posture Estimation Using Neural Networks

    K. Takahashi, T. Uemura and Jun Ohya

    6th International Conference on Engineering Applications of Neural Networks     229 - 236  2000.07  [Refereed]

  • Epipolar constraint from 2D affine lines, and its application in face image rendering

    K Sengupta, J Ohya

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E83D ( 7 ) 1567 - 1573  2000.07  [Refereed]

     View Summary

    This paper has two parts. In the first part of the paper, we note the property that under the para perspective camera projection model of a camera, the set of 2D images produced by a 3D point can be optimally represented by two lines in the affine space (alpha - beta space). The slope of these two lines are same, and we observe that this constraint is exactly the same as the epipolar line constraint. Using this constraint, the equation of the epipolar line can be derived. In the second part of the paper, we use the "same slope" property of the lines in the alpha - beta space to derive the affine structure of the human face. The input to the algorithm is not limited to an image sequence of a human head under rigid motion. It can be snapshots of the human face taken by the same or different cameras, over different periods of time. Since the depth variation of the human face is not very large, we use the para perspective camera projection model. Using this property, we reformulate the (human) face structure reconstruction problem in terms of the much familiar multiple baseline stereo matching problem. Apart from the face modeling aspect. we also show how we use the results for reprojecting human faces in identification tasks.

  • 実時間非接触非装着三次元人物姿勢推定方法に関する一考察

    高橋和彦, 坂口竜己, 大谷 淳

    電子情報通信学会論文誌 D-Ⅱ   J83-D-Ⅱ ( 5 ) 1305 - 1314  2000.05  [Refereed]

  • 人物像の実時間認識・生成

    システム制御情報学会誌 システム/制御/情報   Vol.44 ( No.3 ) 135 - 140  2000.03  [Refereed]

  • 342 A Wavelet-Transform Based Asker Identification System for Smart Multi-point Tele-conferences

    TANG Jinshan, NAKATSU Ryohei, KAWATO Shinjiro, OHYA Jun

    Transactions of Visualization Society of Japan   20 ( 1 ) 303 - 306  2000

     View Summary

    This paper proposes a wavelet-based asker identification system for smart multi-point tele-conferences. Our system is composed of three processors. The first processor is called an "ask" action detector, which is used to decide whether an "ask" action is occurring. If an "ask" action is detected, the input image is sent to the face detector that detects the human face from the complex background. The detected face is fed to a wavelet-based face identification processor to recognize the asker.

    DOI CiNii

  • Two-step approach for real-time eye tracking with a new filtering technique

    S Kawato, J Ohya

    SMC 2000 CONFERENCE PROCEEDINGS: 2000 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN & CYBERNETICS, VOL 1-5     1366 - 1371  2000  [Refereed]

     View Summary

    Head and face detection and eye tracking in realtime are the first steps for head gesture recognition and/or face expression recognition for a human-computer interaction interface. We propose a two-step approach for eye tracking in video streams.
    First, we detect or track a point between the eyes. For this task, we apply a special filter rye proposed before. Once we detect the point between the eyes, it is fairly easy to locate the eyes, which are the two small darkest parts on each side of this point. Because defecting the point between the eves is easier and more stable than directly detecting the eyes, the system can robustly track the eyes
    We implement the system on a SGI O2,workstation. The video image size is 320x240 pixels. The system processes images at seven frames per second in the detection mode, and 13 frames per second in the tracking mode without arty special hardware.

  • Locating human face in a complex background including non-face skin colors

    JS Tang, S Kawato, J Ohya, R Nakatsu

    CISST'2000: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON IMAGING SCIENCE, SYSTEMS, AND TECHNOLOGY, VOLS I AND II     117 - 122  2000  [Refereed]

     View Summary

    This paper proposes a skin color information based face location algorithm that can locate the human face in a complex background including non-face skin color regions, such as hands. Our algorithm is divided into three steps. The first step is to use color components of the input image to segment the skin colors from the complex background. The second step is to detect a curve, which is called the division curve, between the hair region and face region. The last step is to use the division curve to estimate the face location. Our algorithm can be applied to some gesture analysis and understanding systems that combine the gestures of the head and hands.

  • Automatic skin-color distribution extraction for face detection and tracking

    S Kawato, J Ohya

    2000 5TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING PROCEEDINGS, VOLS I-III     1415 - 1418  2000  [Refereed]

     View Summary

    For face detection and tracking in a video sequence, many systems are now applying a technique to extract skin-color areas effaces. In these systems, the skin-color model should be built beforehand. However skin-color changes when the lighting condition varies when an input video camera is changed, when the white-balance of the camera is readjusted, etc.
    In this paper we describe a method to build a skin-color model for operation while a face detection system is running. When it has no skin-color model, it uses the differences between adjacent frames to extract moving objects. Once the system recognizes that the moving object is a face, it analyses the color histograms of the moving urea and extract skin-color distribution
    We applied this method to our face detection and tracking system. Experiments showed good results.

  • Adaptive human motion tracking using non-synchronous multiple viewpoint observations

    A Utsumi, H Yang, J Ohya

    15TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 4, PROCEEDINGS     607 - 610  2000  [Refereed]

     View Summary

    In this paper, we propose and adaptive human tracking system with non-synchronous multiple observations. Our system consists of three types of processes, discovering node for detecting newly appeared person, tracking node for tracking each target person, and observation node for processing one viewpoint (camera) images. We have multiple observation nodes and each node works fully independently. The tracking node integrates observed information based on reliability evaluation. Both observation conditions (number of cameras), relative distance between a human and cameras, extent of occlusion, etc.) and human motion states (walking, standing, sitting) are considered in the evaluation. Matching between tracking models and observed image features are performed in each observation node based on the position, size and color similarities of each 2D image. Due to the non-synchronous property, this system is highly scalable for increasing the detection area and number of observing nodes. Experimental results for some indoor scenes are also described.

    DOI

  • Human body postures from trinocular camera images

    Shoichiro Iwasawa, Jun Ohya, Kazuhiko Takahashi, Tatsumi Sakaguchi, Kazuyuki Ebihara, Shigeo Morishima

    Proceedings - 4th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2000     326 - 331  2000  [Refereed]

     View Summary

    This paper proposes a new real-time method for estimating human postures in 3D from trinocular images. In this method, an upper body orientation detection and a heuristic contour analysis are performed on the human silhouettes extracted from the trinocular images so that representative points such as the top of the head can be located. The major joint positions are estimated based on a genetic algorithm-based learning procedure. 3D coordinates of the representative points and joints are then obtained from the two views by evaluating the appropriateness of the three views. The proposed method implemented on a personal computer runs in real-time. Experimental results show high estimation accuracies and the effectiveness of the view selection process. © 2000 IEEE.

    DOI

    Scopus

    19
    Citation
    (Scopus)
  • Real-time detection of nodding and head-shaking by directly detecting and tracking the "between-eyes"

    Shinjiro Kawato, Jun Ohya

    Proceedings - 4th IEEE International Conference on Automatic Face and Gesture Recognition, FG 2000     40 - 45  2000  [Refereed]

     View Summary

    Among head gestures, nodding and head-shaking are very common and used often. Thus the detection of such gestures is basic to a visual understanding of human responses. However it is difficult to detect them in real-time, because nodding and head-shaking are fairly small and fast head movements. We propose an approach for detecting nodding and head-shaking in real time from a single color video stream by directly detecting and tracking a point between the eyes, or what we call the "between-eyes". Along a circle of a certain radius centered at the "between-eyes", the pixel value has two cycles of bright parts (forehead and nose bridge) and dark parts (eyes and brows). The output of the proposed circle-frequency filter has a local maximum at these characteristic points. To distinguish the true "between-eyes" from similar characteristic points in other face parts, we do a confirmation with eye detection. Once the "between-eyes" is detected, a small area around it is copied as a template and the system enters the tracking mode.combining with the circle-frequency filtering and the template, the tracking is done not by searching around but by selecting candidates using the template
    the template is then updated. Due to this special tracking algorithm, the system can track the "between-eyes" stably and accurately. It runs at 13 frames/s rate without special hardware. By analyzing the movement of the point, we can detect nodding and head-shaking. Some experimental results are shown. © 2000 IEEE.

    DOI

    Scopus

    88
    Citation
    (Scopus)
  • Multiple-Camera-Based Human Tracking using Non-synchronous Observations

    Fourth Asian Conference on Computer Vision     1034 - 1039  2000.01  [Refereed]

  • Neural-network-based real-time human body posture estimation

    K Takahashi, T Uemura, J Ohya

    NEURAL NETWORKS FOR SIGNAL PROCESSING X, VOLS 1 AND 2, PROCEEDINGS     477 - 486  2000  [Refereed]

     View Summary

    This paper proposes a real-time human body posture estimation method using ANNs. The network is composed of three ANNs and a decision logic unit. The ANNs' input is the result of a function analysis on a human silhouette's contour extracted from camera images and the ANNs' output indicates the feature points' positions on the contour. The decision logic unit synthesizes each of the ANNs' output vectors and then the 2D coordinates of the human body's feature points are calculated. The proposed method is implemented on a personal computer and runs in real-time (17-20 frames/sec). Experimental results confirm both the feasibility and the effectiveness of the proposed method for estimating human body postures.

  • Remarks on a real-time 3D human body posture estimation method using trinocular images

    K Takahashi, T Sakaguchi, J Ohya

    15TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 4, PROCEEDINGS   Vol. 4   693 - 697  2000  [Refereed]

     View Summary

    This paper proposes a new real-time method of estimating human postures in 3D from trinocular images. The proposed method extracts feature points Of the human body by applying a type of function analysis to contours of human silhouettes. To overcome self-occlusion problems, dynamic compensation is carried out using the Kalman filter and all feature points are tracked. The 3D coordinates of the feature points are reconstructed by considering the geometrical relationship between the three cameras. Experimental results confirm both the feasibility and the effectiveness of the proposed method, and an application example of the 3D human body posture estimation to a motion recognition system is presented.

  • Virtual Me: A virtual communication method that enables simultaneous multiple existence as an avatar and/or agents

    J Ohya, R Nakatsu, S Kawato, T Sakaguchi

    2000 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, PROCEEDINGS VOLS I-III     1657 - 1660  2000  [Refereed]

     View Summary

    This paper proposes a new communication method, called "Virtual Me", that enables a user to simultaneously participate in multiple Events such as meetings or lectures as a virtual-metamorphosed avatar controlled by the user and/or as a computer-controlled autonomous agents. By interpreting non-verbal data generated by the other participants at each event, the system determines when to switch between using virtual metamorphosis or employing agents. We demonstrate the effectiveness of this method in a prototype of the Virtual Me system. It is our hope that this technology may be in the interest not only of time efficiency but also of enriched human communication.

  • Modeling and Animation of Botanical Tree for Interactive Virtual Environments

    Symposium on Virtual Reality Software and Technology 1999    1999.12  [Refereed]

  • Indexing method for three-dimensional position estimation

    Fermin, I, S Semwal, J Ohya

    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS   E82D ( 12 ) 1597 - 1604  1999.12  [Refereed]

     View Summary

    Indexing techniques usually are used in model-based object recognition and ray tracing algorithms. In this paper we present a new method for estimating the three-dimensional position of a subject (resp. object) in a circumscribed space based on an indexing method. We construct two and three-dimensional indices of a space, which are used to estimate the three-dimensional position by an interpolation technique. There are two processes in estimating the three-dimensional position of a subject (resp. object): preprocessing and three-dimensional position estimation. We have implemented this idea using stereo camera, and tested by using two different sizes of a grid pattern. Promising results for preprocessing and 3D position estimation are presented. Moreover, we show that this approach can also be extended for multiple cameras.

  • A Face Recognition System Based on Wavelet Transform and Neural Network

    International conference on wavelet analysis and its applications    1999.11  [Refereed]

  • A New Size-Limited Filter Bank for Finite-signal Subband Processing

    International conference on wavelet analysis and its applications    1999.11  [Refereed]

  • Face Detection from a Complex Background

    International Workshop on Very Low Bitrate Video Coding    1999.10  [Refereed]

  • Real-time computer vision based technologies for capturing human motions

    14th International Symposium on Computer and Information Sciences    1999.10  [Refereed]

  • Virtual Metamorphosis Systems

    1999 IEEE International Conference on Systems, Man, and Cybernetics    1999.10  [Refereed]

  • Comparison of Neural-Network-Based Pattern Classification Methods with Application to Human Motion Recognition

    Fifth International Conference on Engineering Applications of Neural Networks (EANN '99)    1999.09  [Refereed]

  • 実写映像に基づいた3次元樹木モデルの生成(共著)

    坂口竜己, 大谷 淳, 中津良平

    電子情報通信学会論文誌   J82-D-Ⅱ ( 9 ) 1469 - 1477  1999.09  [Refereed]

  • Towards the realization of non-verbal communication

    HCI'99 (Human-Computer Interaction)   Vol.2  1999.08  [Refereed]

  • 多数カメラを用いた両手手振りの検出

    情報処理学会論文誌   Vol.40, No.8  1999.08  [Refereed]

  • Multiple-Hand-Gesture-Tracking using Multiple Cameras

    IEEE Computer Society Conference on Computer Vision and Pattern Recogntion (CVPR99)    1999.06  [Refereed]

  • Real-time 3-D facial image reconstruction for virtual space teleconferencing

    K Ebihara, N Suzuki, J Ohya, F Kishino

    ELECTRONICS AND COMMUNICATIONS IN JAPAN PART III-FUNDAMENTAL ELECTRONIC SCIENCE   82 ( 5 ) 80 - 90  1999.05  [Refereed]

     View Summary

    In this paper, a new method was proposed for implementing 3-D facial image models designed to allow faithful reconstruction of facial images in virtual space teleconferencing by using 3-D measurement while detecting various facial expressions. In the method proposed, first many dots are painted on the face. Then, for a variety of facial expressions (eight expressions in this study) selected with regard to the action of the major facial muscles, 3-D displacement vectors for the normal (neutral) face are measured at each dot and recorded in the facial image plane as reference vectors. When the facial image is reconstructed, 2-D displacement vectors are detected by tracking markers that were previously placed on the face and are represented as the sum of two enclosing reference vectors. Based on these data, the vertices of a 3-D wireframe face model (WFM) are moved appropriately, and the facial expression is reconstructed. (C) 1999 Scripta Technica.

  • Virtual metamorphosis

    J Ohya, J Kurumisawa, R Nakatsu, K Ebihara, S Iwasawa, D Harwood, T Horprasert

    IEEE MULTIMEDIA   6 ( 2 ) 29 - 39  1999.04  [Refereed]

     View Summary

    The virtual metamorphosis system lets people change their forms into any other form in a virtual scene. To realize these changes, a computer Vision system estimates facial expressions and body postures and reproduces them in a computer graphics avatar in real time. We introduce three systems in order of their development: the Virtual Kabuki system, Networked Theater, and "Shall We Dance?".

    DOI

    Scopus

    20
    Citation
    (Scopus)
  • 多数カメラを用いた手形状認識法とその仮想空間インタフェースへの応用

    内海 章, 大谷 淳, 中津良平

    情報処理学会論文誌   40 ( 2 ) 585 - 593  1999.02  [Refereed]

  • Virtual reality technologies for multimedia communications

    J Ohya, T Miyasato, R Nakatsu

    MIXED REALITY     285 - 300  1999  [Refereed]

  • Extracting facial motion parameters by tracking feature points

    T Otsuka, J Ohya

    ADVANCED MULTIMEDIA CONTENT PROCESSING   1554   433 - 444  1999  [Refereed]

     View Summary

    A method for extracting facial motion parameters is proposed. The method consists of three steps. First, the feature points of the face, selected automatically in the first frame, are tracked in successive frames. Then, the feature points are connected with Delaunay triangulation so that the motion of each point relative to the surrounding points can be computed. Finally, muscle motions are estimated based on motions of the feature points placed near each muscle. The experiments showed that the proposed method can extract facial motion parameters accurately. In addition, the facial motion parameters are used to render a facial animation sequence.

  • Real-time estimation of human body postures using Kalman filter

    K Takahashi, T Sakaguchi, J Ohya

    RO-MAN'99: 8TH IEEE INTERNATIONAL WORKSHOP ON ROBOT AND HUMAN INTERACTION     189 - 194  1999  [Refereed]

     View Summary

    This paper presents a hybrid estimation method of human body postures from CCD camera images. In the hybrid estimation method, the feature points of the human body (top of the head, tips of the hands, and feet, and elbow joints) are obtained from the results of heuristic contour analyses of human silhouettes or those of a time subtraction image depending on the reliability of the silhouette information. A dynamic compensation is then carried out by tracking all feature points using the AR model in order to obtain their optimal position and to overcome self-occlusion problems. The AR model's parameters are estimated through on-line processing by the Kalman filter. The proposed method is implemented on a personal computer and the process runs in real-time. Experimental results show high estimation accuracy and the feasibility of the proposed method.

  • Real-time, 3D estimation of human body postures from trinocular images

    Shoichiro Iwasawa, Jun Ohya, Kazuhiko Takahashi, Tatsumi Sakaguchi, Sinjiro Kawato, Kazuyuki Ebihara, Sigeo Morishima

    Proceedings - IEEE International Workshop on Modelling People, MPeople 1999     3 - 10  1999  [Refereed]

     View Summary

    This paper proposes a new real-time method for estimating human postures in 3D from trinocular images. In this method, an upper body orientation detection and a heuristic contour analysis are performed on the human silhouettes extracted from the trinocular images so that representative points such as the top of the head can be located. The major joint positions are estimated based on a genetic algorithm based learning procedure. 3D coordinates of the representative points and joints are then obtained from the two views by evaluating the appropriateness of the three views. The proposed method implemented on a personal computer runs in real-time (30 frames/second). Experimental results show high estimation accuracies and the effectiveness of the view selection process.

    DOI

    Scopus

    31
    Citation
    (Scopus)
  • A real-time surveillance and evaluation system of the audience's reaction for meeting sessions

    JS Tang, J Ohya

    INTELLIGENT ROBOTS AND COMPUTER VISION XVIII: ALGORITHMS, TECHNIQUES, AND ACTIVE VISION   3837   45 - 57  1999  [Refereed]

     View Summary

    In this paper, we describe a visual surveillance system for evaluating the audience's reaction in meeting sessions. The system, which works in real-time, can recognize and evaluate the reaction of the audience. It is mainly composed of three subsystems. The first subsystem is a face detection and head motion segmentation system which is used to detect the face from complex background and segment the head motion into different units, with each unit including different information about the audience's reaction. The second subsystem is a gesture and pose recognition system which can recognize the gestures and the poses of human head. The third subsystem is an evaluation system which is used to evaluate the reaction of the audience by using the recognition results in the second system. Our system has been tested for determining the concentration of the audience in meeting sessions and the experimental results are good.

  • Artistic anatomy based, real-time reproduction of facial expressions in 3D face models

    J Ohya, K Ebihara, J Kurumisawa

    IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA COMPUTING AND SYSTEMS, PROCEEDINGS VOL 2   Vol.2   684 - 689  1999  [Refereed]

     View Summary

    This paper proposes a new real-time method for reproducing facial expressions in 3D face models realistically based on anatomy for artists. To reproduce facial expressions in a face model, the detected expressions need to be converted to the data for deforming the face model. In the proposed method, an artist who has learned anatomy for artists creates arbitrary facial expressions in the 3D face model by mixing the reference expressions chosen by the artist so that the synthesized expressions realistically represent the respective expressions displayed by real persons. The parameters obtained by this manual operations are used to construct the equations that convert the expression features obtained by the detection module to the displacement vectors of the vertices of the face model. During human communications through face models, the equations are used to reproduce the detected expressions in real-time. The effectiveness and robustness of the proposed method were demonstrated by experimental results and demonstration systems.

  • Novel scene generation, merging and stitching views using the 2D affine space

    K Sengupta, J Ohya

    SIGNAL PROCESSING-IMAGE COMMUNICATION   14 ( 1-2 ) 39 - 53  1998.11  [Refereed]

     View Summary

    In this paper we present a unified theoretical framework for novel scene synthesis, merging real and virtual worlds, and view stitching. To start with, we have a set of real images from weakly calibrated cameras, for which we compute the dense point match correspondences. For applications like novel view synthesis, one may first solve the 3D scene reconstruction problem, followed by a view rendering process. However, errors in 3D scene reconstruction usually gets reflected in the quality of the new scene generated, so we seek. a more direct method. In this paper, we use the knowledge of dense point matches and their affine coordinate values to estimate the corresponding affine coordinate values in the new scene. Our technique of reprojection is extended for other applications like merging real and synthetic worlds, and view stitching. (C) 1998 Elsevier Science B.V. All rights reserved.

    DOI

  • Face/gesture analysis/synthesis telchnologies for humanto-human communications through virtual environments

    Jun Ohya

    Proceedings of the 6th ACM International Conference on Multimedia: Face/Gesture Recognition and their Applications, MULTIMEDIA 1998     12 - 19  1998.09  [Refereed]

     View Summary

    This paper describes the vutual metamorphosis system, whtch was originally proposed by ATR Media Integratton &amp
    Communicattons Research Laboratories, Kyoto, Japan. In the vntual metamorphosis system, people can change their forms mto any other form in a vutual scene. To realize. this changes, facial expressions and body postures are estimated by computer vision and reproduced m a computer graphics avatar m real-ume. We developed computer vision based technologies for automatic face trackmg, estimatmg deformations of factal components,and estimatmg body postures in real-time. We also developed real-time facial expresston reproduction method based on "Anatomy for artists". Usmg these technologies, prototyping systems for virtual metamorpho-SIS were constructed. This paper introduces two systems in order of their development: the Virtual Kabuki system and the "Shall We Dance" system.

    DOI

    Scopus

  • 画素値分布の推定による時系列画像からの動物体領域抽出

    内海 章, 大谷 淳, 中津良平

    電子情報通信学会論文誌 D-II   J81-D-II ( 8 ) 1768 - 1775  1998.08  [Refereed]

  • 美術解剖学に基づいた仮想変身システムのための実時間人物表情再現(共著)

    海老原一之, 楜沢 順, 大谷 淳, 中津良平, 岸野文郎

    電子情報通信学会論文誌   J81-D-Ⅱ ( 5 ) 841 - 849  1998.05  [Refereed]

    CiNii

  • 複雑な輪郭の自動抽出・追跡法

    畑幸一, 大谷 淳, 岸野文郎, 中津良平

    電子情報通信学会論文誌 D-II   .J81-D-II ( 4 ) 706 - 715  1998.04  [Refereed]

  • A new robust real-time method for extracting human silhouettes from color images

    Masanori Yamada, Kazuyuki Ebihara, Jun Ohya

    Proceedings - 3rd IEEE International Conference on Automatic Face and Gesture Recognition, FG 1998     528 - 533  1998  [Refereed]

     View Summary

    A new method for extracting human silhouettes from color images in real-time regardless of changes in lighting conditions and backgrounds is proposed. In the method, RGB values of each pixel in color images acquired by a CCD camera are converted to YIQ values. Before extracting human silhouettes, the mean YIQ values of each pixel and the threshold values for the Y, I, Q components are obtained in advance from the image sequence acquired by the camera that observed the background. Based on the mean values and the threshold values, each pixel is judged whether it belongs to the human silhouette or not. The proposed method was used with the virtual metamorphosis system, in which anyone can change his or her form to any other form, and the robustness and real-time-ness were confirmed. © 1998 IEEE.

    DOI

    Scopus

    28
    Citation
    (Scopus)
  • Spotting segments displaying facial expression from image sequences using HMM

    Takahiro Otsuka, Jun Ohya

    Proceedings - 3rd IEEE International Conference on Automatic Face and Gesture Recognition, FG 1998     442 - 447  1998  [Refereed]

     View Summary

    A method that can be used for spotting segments that display facial expression is proposed. The motion of the face is modeled by HMM in such a way that each state corresponds to the conditions of facial muscles, e.g., relaxed, contracting, apex and relaxing. The probability assigned to each state is updated iteratively as the feature vector is obtained from image processing. A spotted segment is placed into a certain category when the probability of that category exceeds a threshold value. Experiments show that the segments for the six basic expressions can be spotted accurately in near real time. © 1998 IEEE.

    DOI

    Scopus

    47
    Citation
    (Scopus)
  • A new camera projection model and its application in reprojection

    K Sengupta, J Ohya

    FOURTEENTH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOLS 1 AND 2     59 - 63  1998  [Refereed]

     View Summary

    In this paper we present a new camera projection model, which is intermediate between the affine camera mode and the pin hole projection model. It is modeled as a perspective projection of 3D points into an arbitrary plane, followed by an affine transform of these projected points. We observe that the reprojection of a point into a navel image can be achieved uniquely provided that we have located a set of Jive reference points over four images (of which three are input images, and the fourth is the novel image). Also, the reprojection, theory; does not assume that the input images are captured from cameras with identical internal calibration parameters. Thus, we apply our technique two different domain: (a) generation of novel images from a stereo pair (b) generation of virtual walkthrough sequence with a monocular image sequence as input.

  • Geometric-imprints: A significant points extraction method for the scan & track virtual environment

    SK Semwal, J Ohya

    AUTOMATIC FACE AND GESTURE RECOGNITION - THIRD IEEE INTERNATIONAL CONFERENCE PROCEEDINGS     480 - 485  1998  [Refereed]

     View Summary

    We have implemented a new algorithm, called the geometric-imprints method for extracting key-features of a given contour. Series of points on the curve are used to find a geometric-imprint set of the contour. The algorithm is based upon the assumption that an articulated human and other figures have mostly cylindrical body-parts. The algorithm recursively partitions the contour based upon the cylindrical nature of human-silhouettes, and a starting point. Similar topological contours are expected to produce similar geometric-imprints. Thus the correspondence problem is somewhat simplified. Results of our implementation are presented.

  • Human face structure estimation from multiple images using the 2D affine space

    K Sengupta, J Ohya

    AUTOMATIC FACE AND GESTURE RECOGNITION - THIRD IEEE INTERNATIONAL CONFERENCE PROCEEDINGS     106 - 111  1998  [Refereed]

     View Summary

    In this paper we present an algorithm to estimate the human face structure. The input to the algorithm is not limited to an image sequence of a human head under rigid motion. It can be snapshots of the human face taken by the same or different cameras, over different periods of time. Since the depth variation of the human face is not very large, we use the affine camera projection model. Under this assumption, it can be shown that the set of 2D images produced by a 3D point feature of a rigid object can be optimally represented by two lines in the affine space. Using this property, we reformulate the (human) face structure reconstruction problem in terms of the much familiar multiple baseline stereo matching problem [8]. Apart from the face modeling aspect, we also show how we use the results for reprojecting human faces in identification tasks.

  • Generating virtual environments for human communications - Virtual metamorphosis system and novel view generation

    J Ohya, K Sengupta

    COMPUTER VISION FOR VIRTUAL REALITY BASED HUMAN COMMUNICATIONS - 1998 IEEE AND ATR WORKSHOP PROCEEDINGS     43 - 50  1998  [Refereed]

     View Summary

    At ATR Media Integration & Communications Research Laboratories, our research group is aiming at realizing virtual communication environments in which the users at remote sites can feel that they are virtually co-located. This paper presents the ongoing research in two subareas chosen to address this problem: (a) Generating human images in virtual scenes (b) Act image based approach for generation of (background) scenes. In the virtual metamorphosis system, 3D models of characters are created in advance. For metamorphosis, the facial expressions and body posture of a person are detected in real-time fr om the face images and monocular thermal images of the person, respectively. The detected expressions and body movements are reproduced by deforming the 3D models. Demonstrations using SD models of a Kabuki actor and a dinosaur show good performance. for novel view generation, we use as affine coordinate based re-projection scheme, which is also extended to merging real and synthetic objects.

  • Real-time human posture estimation using monocular thermal images

    S Iwasawa, K Ebihara, J Ohya, S Morishima

    AUTOMATIC FACE AND GESTURE RECOGNITION - THIRD IEEE INTERNATIONAL CONFERENCE PROCEEDINGS     492 - 497  1998  [Refereed]

     View Summary

    This paper introduces a new real-lime method to estimate the posture of a human from thermal images acquired by an infrared camera regardless of the background and lighting conditions. Distance transformation is performed for the human body area extracted from the thresholded thermal image for the calculation of the center of gravity. After the orientation of the zipper half of the body is obtained by calculating the moment of inertia, significant points such as the top of the head, the tips oft he hands and foot are heuristically located. In addition, the elbow and knee positions are estimated from the detected (significant) points using a genetic algorithm based learning procedure.
    The experimental results demonstrate the robustness of the proposed algorithm and real-time (faster than 20 frames per second) performance.

  • Multiple-view-based tracking of multiple humans.

    Akira Utsumi, Hiroki Mori, Jun Ohya, Masahiko Yachida

    Fourteenth International Conference on Pattern Recognition, ICPR 1998, Brisbane, Australia, 16-20 August, 1998     597 - 601  1998  [Refereed]

    DOI CiNii

  • Direct Manipulation Interface Using Multiple Cameras for Hand Gesture Recognition.

    Akira Utsumi, Jun Ohya

    IEEE International Conference on Multimedia Computing and Systems, ICMCS 1998, Austin, Texas, USA, June 28 - July 1, 1998     264 - 267  1998  [Refereed]

    DOI

  • Image segmentation for human tracking using sequential-image-based hierarchical adaptation

    Akira Utsumi, Jun Ohya

    Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition     911 - 916  1998  [Refereed]

     View Summary

    We propose a novel method of extracting a moving object region from each frame in a series of images regardless of complex, changing background using statistical knowledge about the target. In vision systems for 'real worlds' like a human motion tracker, a priori knowledge about the target and environment is often limited (e.g., only the approximate size of the target is known) and is insufficient for extracting the target motion directly. In our approach, information about both target object and environment is extracted with a small amount of given knowledge about the target object. Pixel value (color, intensity, etc.) distributions for both the target object and background region are adaptively estimated from the input image sequence based on the knowledge. Then, the probability of each pixel being associated with the target object is calculated. The target motion can be extracted from the calculated stochastic image. We confirmed the stability of this approach through experiments.

    DOI

    Scopus

    11
    Citation
    (Scopus)
  • Multiple Camera Based Human Motion Estimation.

    Akira Utsumi, Hiroki Mori, Jun Ohya, Masahiko Yachida

    Computer Vision - ACCV'98, Third Asian Conference on Computer Vision, Hong Kong, China, January 8-10, 1998, Proceedings, Volume II     655 - 662  1998  [Refereed]

    DOI

    Scopus

  • 距離変換処理を用いた多視点画像による手姿勢推定法(共著)

    内海 章, 宮里 勉, 岸野文郎, 大谷 淳, 中津良平

    映像情報メディア学会誌   51 ( 12 ) 2116 - 2125  1997.12  [Refereed]

     View Summary

    We describe a method for detecting hand position, posture, and finger bendings using multiple camera images. Stable detection can be achieved using distance transformed images. We detect the maximum point in each distance transformed image as the center of gravity (COG) point of the hand region and calculate its 3D position by stereo matching. The distance value of a COG point varies according to the angle between the camera axis and normal axis of the hand plane. Hand rotation angle can be determined in maximum likelihood estimation from the distance values in all camera images. Using the detected position and posture, the best camera for hand shape detection can be selected. This camera selection makes the hand shape detection simple and stable. This system can be used as a user interface device in a virtual environment, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices.

    DOI CiNii

    Scopus

    3
    Citation
    (Scopus)
  • 遺伝的アルゴリズムを用いた顔画像からの表情と顔の姿勢の推定

    大谷 淳

    電子情報通信学会論文誌   J80-D-Ⅱ ( 8 ) 2152 - 2160  1997.08  [Refereed]

  • 連続出力確率密度分布を用いたHMMによる動画像からの複数人物の表情認識

    大塚尚宏, 大谷 淳, 中津良平

    電子情報通信学会論文誌   J80-D-Ⅱ ( 8 ) 2129 - 2137  1997.08  [Refereed]

  • Real - time Estimation of Human Body Postures

    YAMADA Masanori, EBIHARA Kazuyuki, OHYA Jun

    ITE Technical Report   21 ( 42 ) 61 - 66  1997.07

     View Summary

    In our "Virtual KABUKI System", to estimate human body postures, a human silhouette must be segmented from a background image. We had used an infrared camera to segment it. But clothes which are hard to have a heat, such as a skirt and a jacket, obstruct a detection of a complete human silhouette. To solve this problem, we developed the new method which uses a difference between a current image and a background image. This paper introduce a new real-time method to estimate human body postures from color images acquired by a normal CCD camera.

    DOI CiNii

  • Virtual KABUKI System

    EBIHARA Kazuyuki, KURUMISAWA Jun, IWASAWA Syoichiro, OHYA Jun

    ITE Technical Report   21 ( 6 ) 61 - 68  1997.02

     View Summary

    This paper describes the Virtual Kabuki Theater the authors have recently developed. In the Virtual Kabuki Theater, people at different locations can be Kabuki actors in a Kabuki scene. In our system, Kabuki actors'3D models are created in advance. Facial expressions of a person are detected in real-time in the face images from the small camera fixed to the helmet worn by the person. Body movements of the person are estimated in real-time from the thermal images acquired by the infrared camera that observes the person. The detected expressions and body movements are reproduced in the Kabuki actor's model. Our implementation shows good performance. The Virtual Kabuki Theater is a first step towards human metamorphosis systems, in which anyone can change (metamorphose) his/her form into any other characters.

    DOI CiNii

  • Real-time estimation of human body posture from monocular thermal images

    S Iwasawa, K Ebihara, J Ohya, S Morishima

    1997 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS     15 - 20  1997  [Refereed]

     View Summary

    This paper introduces a new;real-time method to estimate the posture of a human from thermal images acquired by an infrared camera regardless of the background and lighting conditions. Distance transformation is performed for the human body area extracted from the thresholded thermal image for the calculation of the center of gravity. After the orientation of the upper half of the body is obtained by calculating the moment of inertia, significant points such as the top of the head, the tips of the hands and foot are heuristically located. In addition, the elbow and knee positions are estimated from the detected (significant) points using a genetic algorithm based learning procedure.
    The experimental results demonstrate the robustness of the proposed algorithm and rear-time (faster than 20 frames per second) performance.

    DOI

  • A Study of Facial Expression Recognition in Real-time Using Optical Flow

    Otsuka Takahiro, Ohya Jun

    PROCEEDINGS OF THE ITE ANNUAL CONVENTION   1997   375 - 376  1997

     View Summary

    A method for recognizing facial expressions in which a feature vector is obtained from Fiourier transform coefficients of velocity vector on a face while an expressions is displayed. As the feature vector is computed from the gradient based optical flow algorithm without time-consuming iteration, the method makes a real-time processing possible.

    DOI CiNii

  • An affine coordinate based algorithm for reprojecting the human face for identification tasks

    K Sengupta, J Ohya

    INTERNATIONAL CONFERENCE ON IMAGE PROCESSING - PROCEEDINGS, VOL III     340 - 343  1997  [Refereed]

     View Summary

    In this paper we present an algorithm to generate new views of a human face, starting with at least two other views of the face. In a typical face recognition system, the task of comparison becomes easier if the faces have similar orientation with respect to the camera. The affine coordinate based reprojection algorithm presented in the paper enables us to do that. Dense point matches between the two input faces of the same individual are computed using an affine coordinate based reprojection framework. This is followed by the reprojection of one of these to faces to the target face once the user has matched four feature points across two input face images and the target face image.

  • Novel scene generation, merging and stitching views using the 2D affine space

    K Sengupta, J Ohya

    IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA COMPUTING AND SYSTEMS '97, PROCEEDINGS     602 - 603  1997  [Refereed]

     View Summary

    In this paper we present an algorithm to generate new views of a scene, starting from stereo images. Errors in 3D scene reconstruction usually get reflected in the quality of the new scene generated, so we seek a direct method for reprojection. In this paper, we use the knowledge of dense point matches and their affine coordinate values to estimate the corresponding affine coordinate values in the new scene.

  • Hand Image Segmentation Using Sequential-Image-Based Hierarchical Adaptation.

    Akira Utsumi, Jun Ohya

    Proceedings 1997 International Conference on Image Processing, ICIP '97, Santa Barbara, California, USA, October 26-29, 1997     208 - 211  1997  [Refereed]

    DOI

  • Real-time estimation of human body postures from monocular thermal images

    Shoichiro Iwasawa, Kazuyuki Ebihara, Jun Ohya, Ryohei Nakatsu, Shigeo Morishima

    Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers   51 ( 8 ) 1270 - 1277  1997  [Refereed]

     View Summary

    This paper proposes a new real-time method for estimating human body postures from thermal images acquired by an infrared camera, regardless of the background and lighting conditions. Distance transformation is performed for the human body area extracted from the thresholded thermal image, in order to calculate the center of gravity. After the orientation of the upper half of the body is obtained by calculating the moment of inertia, significant points such as the top of the head and the ends of the hands and feet are heuristically located. In addition, the elbow and knee positions are estimated from the detected (significant) points, using a genetic-algorithm-based learning procedure. The experimental results demonstrate the robustness of the proposed algorithm and real-time performance (faster than 20 frames per second).

    DOI

    Scopus

    4
    Citation
    (Scopus)
  • Model-based estimation of human posture parameters from multiple camera images using genetic algorithms

    Jun Ohya, Fumio Kishino

    Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers   51 ( 12 ) 2107 - 2115  1997  [Refereed]

     View Summary

    A new method for estimating human posture from multiple images using a genetic algorithm is proposed. In the proposed algorithm, the posture parameters to be estimated are assigned to the genes of an individual in the population. For each individual, its fitness evaluates to what extent the human multiple images synthesized by deforming a 3D human model according to the values of the genes are registered to the real human multiple images. Genetic operations such as natural selection, crossover and mutation are performed so that individuals in the next generation are generated. After a certain number of repetitions of these processes, the estimated parameter values are obtained from the individual with the best fitness. Experiments using multiple synthesized images show promising results for estimating 17 joint angle values for each degree of freedom of the joints and also the three translational and three rotational degrees of freedom.

    DOI

    Scopus

  • 臨場感通信会議のための実時間表情検出

    海老原一之, 大谷 淳, 岸野文郎

    テレビジョン学会誌   50 ( 10 ) 1497 - 1506  1996.10  [Refereed]

     View Summary

    A new method for the real-time detection of facial expressions from time-sequential images is proposed. The proposed method does not require the use of tape marks that are pasted to the face to detect expressions in real-time in the current implementation for virtual space teleconferencing. In the proposed method, four windows are applied to four areas of a facial image : left and right eyes, mouth and forehead. Each window is divided into blocks consisting of 8 by 8 pixels. Discrete cosine transform (DCT) is applied to each block, and the feature vector of each window is obtained by taking the summations of the DCT energies in the horizontal, vertical and diagonal directions. To convert the DCT features to virtual tape mark movements, we represent the displacement of a virtual tape mark by a polynomial of the DCT features for the three directions. We apply a genetic algorithm to train facial expression image sequences to find the optimal set of coefficients that minimizes the difference between the real and converted displacements of the virtual tape marks. Experimental results shows the effectiveness of the proposed method.

    DOI CiNii

  • 形状の面対称性に基づく能動的観測による3次元物体形状の復元および色彩情報の再現

    吉田和憲, 大谷 淳, 岸野文郎, 田中弘美

    テレビジョン学会誌   50 ( 10 ) 1536 - 1542  1996.10  [Refereed]

     View Summary

    We propose a shape reconstruction and color inferring method for 3D objects based on planar symmetry using an active camera. This efficiently creates 3-D models of objects from the binary silhouettes in some views acquired by the active camera whose viewpoints are selected recursively based on the symmetry planes of observed objects. And the lighting condition is measured using color samples before inferring the color of observed objects. The original color data of the plane is inferred from the observed RGB data of the color samples, RGB data of the object observed by the CCD camera, and the surface normal to the point on the surface of the object.

    DOI CiNii

  • 臨場感通信会議のための実時間3次元表情再現

    海老原一之, 鈴木紀子, 大谷 淳, 岸野文郎

    電子情報通信学会論文誌A   J79-A ( 2 ) 527 - 536  1996.02  [Refereed]

  • Detecting facial expressions from face images using a genetic algorithm

    Jun Ohya, Fumio Kishino

    Proceedings - International Conference on Pattern Recognition   3   649 - 653  1996  [Refereed]

     View Summary

    A new method to detect deformations of facial parts from a face image regardless of changes in the position and orientation of a face using the genetic algorithm is proposed. Facial expression parameters that are used to deform and position a 3D face model are assigned to the genes of an individual in a population. The face model is deformed and positioned according to the gene values of each individual and is observed by a virtual camera, and a face image is synthesized. The fitness which evaluates to what extent the real and synthesized face images are similar to each other is calculated. After this process is repeated for sufficient generations, the parameter estimation is obtained from the genes of the individual with the best fitness. Experimental results demonstrate the effectiveness of the method. © 1996 IEEE.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Automatic extraction and tracking of contours

    Koichi Hata, Jun Ohya, Fumio Kishino, Ryohei Nakatsu

    Proceedings - International Conference on Pattern Recognition   1   441 - 445  1996  [Refereed]

     View Summary

    This paper considers the problem of extracting and tracking complex contours without user interaction. We assume that a complex contour consists of contour segments whose spatial coordinates and intensity gradient vary smoothly in the direction normal to themselves. In our algorithm, digital curves that could correspond to contour segments are extracted by connecting edge pixels using a B-spline based contour segment model. The extracted curves trace the contour segments at the next frame by using the active contour model technique. Experimental results show even occluded contours can be tracked automatically. © 1996 IEEE.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Real-time facial expression detection based on frequency domain transform

    K Ebihara, J Ohya, F Kishino

    VISUAL COMMUNICATIONS AND IMAGE PROCESSING '96   2727   916 - 926  1996  [Refereed]

  • Virtual Kabuki Theater: Towards the realization of human metamorphosis systems

    J Ohya, K Ebihara, J Kurumisawa, R Nakatsu

    RO-MAN '96 - 5TH IEEE INTERNATIONAL WORKSHOP ON ROBOT AND HUMAN COMMUNICATION, PROCEEDINGS     416 - 421  1996  [Refereed]

     View Summary

    This paper describes the Virtual Kabuki Theater the authors have recently developed. In the Virtual Kabuki Theater, people at different locations can be Kabuki actors in a Kabuki scene. In our system, Kabuki actors' SD models are created in advance. Facial expressions of a person are detected an real-tame in the face images from the small camera fixed to the helmet worn by the person. Body movements of the person are estimated in real-time from the thermal images acquired by the infrared camera that observes the person. The detected expressions and body movements are reproduced in. the Kabuki actor's model. Our implementation shows good performance. The Virtual Kabuki Theater is a first step towards human metamorphosis systems, in which anyone can change (metamorphose) his/her form into any other characters.

  • Pose estimation of human body part using multiple cameras

    K Sengupta, J Ohya

    RO-MAN '96 - 5TH IEEE INTERNATIONAL WORKSHOP ON ROBOT AND HUMAN COMMUNICATION, PROCEEDINGS     146 - 151  1996  [Refereed]

     View Summary

    In this paper, we present a method of obtaining the approximate transformation parameter values as a starting point in estimating the pose of rigid 3D free form objects using multiple 2D images. We back project the edge silhouettes in the images, and obtain the approximate volume in the 3D space containing the object. Next, for a point selected in the volume, we hypothesize a set of points within the 3D CAD model of the object it can possibly correspond to, using the spatial extent function introduced in this paper. This is repeated for three arbitrarily chosen point in the volume. The hypothesized (match point) lists of these three points are next used to derive the pose parameter by enforcing the conditions of rigidity. Our initial experiments demonstrate the potential of this idea, and the pose parameters estimated using this method can be refined using the standard methods available in the literature.

  • Realistic and real-time reproduction of human body movements for virtual space teleconferencing

    Keiichi Nariyama, Karansher Singh, Jun Ohya, Fumio Kishino

    Terebijon Gakkaishi (Journal of the Institute of Television Engineers of Japan)   50 ( 9 ) 1352 - 1362  1996  [Refereed]

     View Summary

    This paper describes realistic, real-time 3-D reproduction of human body movements based on detection data from a small number of sensors attached to a participant for virtual space teleconferencing. A 3-D human model is created by a skeletal link model. WFM (Wire Frame Model), body segments for FFD (Free Form Deformations), and IFP (Implicit Function Primitives) for body parts. In this model WFM provides fast rendering ability and FFD and IFP give natural deformation representation. A reconstruction algorithm that can deal with human movements in teleconference situations is proposed
    sensed data are used to reconstruct whole body movements by utilizing motion constraints. Experiments, in which four magnetic sensors for detecting the six pose parameters are attached to the upper half of a human body, show the effectiveness of the proposed 3-D human model and reconstruction method, in which human movements are reproduced at a speed of 8 frames/sec.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • 隠れマルコフモデルによる顔動画像からの表情認識

    坂口竜己, 大谷 淳, 岸野文郎

    テレビジョン学会誌   49 ( 8 ) 1060 - 1067  1995.08  [Refereed]

     View Summary

    A method for recognizing facial expressions from time-sequential images by using Hidden Markov Models (HMM) is proposed. HMM has the advantage that it can process time-sequential infomation. Moreover we can expect the HMM to make generalizations from the training data because of its learning procedure. Each image of a facial expression is transformed into an image feature vector. Each element of the feature vector consists of the average power from a distinct frequency band obtained by applying the Wavelet transformation to the image. The sequence is converted into a symbol sequence by using a new category-separated vector quantization. The codebook is constructed by appending codewords selected from other categories to each category to reduce the probability of wrong symbolization for similar facial expressions. To recognize an observed sequence, the HMM that best matches the sequence is chosen, and the category of the HMM is the recognized expression. Experiments for recognizing 4 expressions result in a promising recognition rate of 93.7%.

    DOI CiNii

  • VIRTUAL SPACE TELECONFERENCING - REAL-TIME REPRODUCTION OF 3D HUMAN IMAGES

    J OHYA, Y KITAMURA, F KISHINO, N TERASHIMA, H TAKEMURA, H ISHII

    JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION   6 ( 1 ) 1 - 25  1995.03  [Refereed]

     View Summary

    Real-time reproduction of 3D human images is realized by an experimental system recently built as a prototype for virtual space teleconferencing, in which participants at different sites can feel as if they are colocated and can work cooperatively, At each sending and receiving site of the teleconferencing system, a 3D model of each participant is constructed from a wire frame model mapped by color texture and is rendered on a 3D display, In the current experimental system, real-time detection of facial features at the sending site is achieved by visually tracking tape marks pasted to the participant's face. Movements of the head, body, hands, and fingers are detected in real time using magnetic sensors and data gloves. At the receiving site, the detected motion parameters are used to move nodes in the wire frame model to reproduce the movements of the participants at each sending site. Realistic facial expressions are reproduced through the-use of simple motion rules applied to the tape mark tracking information, Through experimental evaluation, the optimal number of nodes for best quality has been obtained. Reproduction of facial expressions and synthesis of arm movements are examined, The reproduction speed using the optimal human model is approximately 6 frames/s. Examples of cooperative work by participants using the experimental system illustrate the effectiveness of virtual space teleconferencing. (C) 1995 Academic Press, Inc.

    DOI

    Scopus

    38
    Citation
    (Scopus)
  • Real Time Facial Expression Detection Based On Frequency Domain Transform

    EBIHARA Kazuyuki, OHYA Jun, KISHINO Fumio

    Proceedings of The ITE Winter Annual Convention   1995   82 - 82  1995

    DOI CiNii

  • A study of real time facial expression detection for virtual space teleconferencing

    K Ebihara, J Ohya, F Kishino

    RO-MAN'95 TOKYO: 4TH IEEE INTERNATIONAL WORKSHOP ON ROBOT AND HUMAN COMMUNICATION, PROCEEDINGS     247 - 252  1995  [Refereed]

  • Realistic 3D facial animation in virtual space teleconferencing

    L Moubaraki, J Ohya, F Kishino

    RO-MAN'95 TOKYO: 4TH IEEE INTERNATIONAL WORKSHOP ON ROBOT AND HUMAN COMMUNICATION, PROCEEDINGS     253 - 258  1995  [Refereed]

  • Active 3D modeling by recursive viewpoint selection based on symmetry

    K Yoshida, HT Tanaka, J Ohya, F Kishino

    INTELLIGENT ROBOTS AND COMPUTER VISION XIV: ALGORITHMS, TECHNIQUES, ACTIVE VISION, AND MATERIALS HANDLING   2588   326 - 336  1995  [Refereed]

  • RECOGNIZING CHARACTERS IN SCENE IMAGES

    J OHYA, A SHIO, S AKAMATSU

    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE   16 ( 2 ) 214 - 220  1994.02  [Refereed]

     View Summary

    An effective algorithm for character recognition in scene images is studied. Scene images are segmented into regions by an image segmentation method based on adaptive thresholding. Character candidate regions are detected by observing gray-level differences between adjacent regions. To ensure extraction of multisegment characters as well as single-segment characters, character pattern candidates are obtained by associating the detected regions according to their positions and gray levels. A character recognition process selects patterns with high similarities by calculating the similarities between character pattern candidates and the standard patterns in a dictionary and then comparing the similarities to the thresholds. A relaxational approach to determine character patterns updates the similarities by evaluating the interactions between categories of patterns, and finally character patterns and their recognition results are obtained. Highly promising experimental results have been obtained using the method on 100 images involving characters of different sizes and formats under uncontrolled lighting.

    DOI

    Scopus

    177
    Citation
    (Scopus)
  • DENSE, TIME-VARYING RANGE DATA-ACQUISITION FROM STEREO PAIRS OF THERMAL AND INTENSITY IMAGES

    J OHYA, F KISHINO

    1994 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS     860 - 865  1994  [Refereed]

  • HUMAN POSTURE ESTIMATION FROM MULTIPLE IMAGES USING GENETIC ALGORITHM

    J OHYA, F KISHINO

    PROCEEDINGS OF THE 12TH IAPR INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION - CONFERENCE A: COMPUTER VISION & IMAGE PROCESSING     750 - 753  1994  [Refereed]

  • 隠れマルコフモデルを用いた動画像からの人物の行動認識(共著)

    大和淳司, 大谷 淳, 石井健一郎

    電子情報通信学会論文誌   J76-D-Ⅱ ( 12 ) 2556 - 2563  1993.12  [Refereed]

    CiNii

  • 3次元形状の軸表現:Smoothed Local Generalized Cones

    佐藤嘉伸, 大谷 淳, 石井健一郎

    電子情報通信学会論文誌D-II   J76-D-II ( 8 ) 1612 - 1623  1993.08  [Refereed]

  • 距離画像からの3次元形状の階層的パーツ構造の復元

    佐藤嘉伸, 大谷 淳, 石井健一郎

    電子情報通信学会論文誌D-II   J76-D-II ( 8 ) 1624 - 1636  1993.08  [Refereed]

  • Perceptual Kinematics : Vision-based Control of Robot Manipulators

    CUCKA Peter, OHYA Jun, KISHINO Fumio

    ITE Technical Report   17 ( 76 ) 29 - 36  1993

     View Summary

    We present a new method for the control of a robot manipulator. The method assumes virtually no prior Information about the robot's kinematic behavior and therefore avoids the costly and time-consuming calibration required by many existing methods. Instead, it relies on continuous sensory feedback, integrating visual data and kinematics into a single mapping, called the perceptual kinematic map (PKM), from the robot's joint space to the camera space. We present results from a real-time, software simulation of a manipulator using a PKM controller.

    DOI CiNii

  • Time-Varying Homotopy and the Animation of Facial Expressions for 3D Human Face Images

    KAJIWARA Souichi, TANAKA Hiromi T., OHYA Jun, KISHINO Fumio

    ITE Technical Report   17 ( 58 ) 37 - 42  1993

     View Summary

    A homotopy describes the transformation of one arbitrary curve into another that shares the same endpoints. In this paper, we propose a deformable cylinder model, based on homotopy, in which an arbitrary surface interpolated between two contours via a blending function is transformed into another surface over time. We then show how this homotopic deformation can be applied to the realistic animation of human faces in a virtual space teleconferencing system. Specifically, we show that facial expressions such as wrinkling of the forehead and opening and closing of the mouth can be synthesized and animated in real time through 3D homotopic deformations.

    DOI CiNii

  • A NEW METHOD FOR ACQUIRING TIME-SEQUENTIAL RANGE IMAGES BY INTEGRATING STEREO PAIRS OF THERMAL AND INTENSITY IMAGES

    J OHYA, F KISHINO

    1993 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION : PROCEEDINGS     730 - 731  1993  [Refereed]

  • VIRTUAL SPACE TELECONFERENCING SYSTEM - REAL-TIME DETECTION AND REPRODUCTION OF 3-D HUMAN IMAGES

    F KISHINO, J OHYA, H TAKEMURA, N TERASHIMA

    HUMAN-COMPUTER INTERACTION, VOL 2   19   669 - 674  1993  [Refereed]

  • REAL-TIME REPRODUCTION OF 3D HUMAN IMAGES IN VIRTUAL SPACE TELECONFERENCING

    J OHYA, Y KITAMURA, H TAKEMURA, F KISHINO, N TERASHIMA

    IEEE VIRTUAL REALITY ANNUAL INTERNATIONAL SYMPOSIUM     408 - 414  1993  [Refereed]

  • TIME-VARYING HOMOTOPY AND THE ANIMATION OF FACIAL EXPRESSIONS FOR 3D VIRTUAL SPACE TELECONFERENCING

    S KAJIWARA, HT TANAKA, Y KITAMURA, J OHYA, F KISHINO

    VISUAL COMMUNICATIONS AND IMAGE PROCESSING 93, PTS 1-3   2094   37 - 48  1993  [Refereed]

  • 多次元心理空間を用いる類似画像検索法

    田邊勝義, 大谷 淳, 石井健一郎

    電子情報通信学会論文誌D-II   J75-D-II ( 11 ) 1856 - 1865  1992.11  [Refereed]

    CiNii

  • FACIAL IMAGE SYNTHESIS BY HIERARCHICAL WIRE FRAME MODEL

    Y KITAMURA, Y NAGASHIMA, J OHYA, F KISHINO

    VISUAL COMMUNICATIONS AND IMAGE PROCESSING 92, PTS 1-3   1818   1358 - 1365  1992  [Refereed]

  • 情景画像からの文字パターンの抽出と認識

    大谷 淳, 塩 昭夫

    電子情報通信学会論文誌D   J71-D ( 6 ) 1037 - 1047  1988.06  [Refereed]

  • リングパターン法を用いるロボット用物体認識センサ

    大谷 淳

    電子通信学会論文誌D   J69-D ( 3 ) 269 - 278  1986.03  [Refereed]

  • GRAY-SCALE PRINTING ON PLAIN PAPER USING THERMAL INK-TRANSFER IMAGING

    J OHYA, Y TOKUNAGA

    JOURNAL OF IMAGING TECHNOLOGY   10 ( 2 ) 57 - 63  1984  [Refereed]

  • COLOR THERMAL INK-TRANSFER IMAGING

    J OHYA, Y TOKUNAGA

    IEEE TRANSACTIONS ON ELECTRON DEVICES   31 ( 9 ) 1258 - 1263  1984  [Refereed]

    DOI

    Scopus

    1
    Citation
    (Scopus)
  • NEW GRAY-SCALE PRINTING METHOD USING A THERMAL PRINTER

    Y TOKUNAGA, K KUBOTA, J OHYA

    IEEE TRANSACTIONS ON ELECTRON DEVICES   30 ( 8 ) 898 - 904  1983  [Refereed]

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Gray Scale Recording on Plain Paper using 3-L Method

    TOKUNAGA Yukio, OHYA Jun, KUBOTA Kazunari

    ITE Technical Report   5 ( 15 ) 45 - 52  1981

    DOI CiNii

▼display all

Books and Other Publications

  • Analyzing video sequences of multiple humans: Tracking, posture estimation and behavior recognition

    Kluwer Academic Publishers  2002.03

Misc

  • 人共存型自律移動ロボットにおけるステレオカメラと機械学習を用いた周辺人物の状態認識および行動予測

    林正晃, 大谷淳, 大和淳司, 亀崎允啓, 斎藤恭介, 濱田太郎, 櫻井絵梨子, 菅野重樹

    電子情報通信学会技術研究報告(Web)   121 ( 420(IMQ2021 10-69) )  2022

    J-GLOBAL

  • 深層学習を用いた心電図からの大動脈弁狭窄症の識別法の提案とGrad-CAMを用いた分析

    秦絵里香, 瀬尾燦振, 中山雅文, 岩崎清隆, 大川内隆朗, 大谷淳

    電子情報通信学会技術研究報告   119 ( 399(MI2019 65-123)(Web) )  2020

    J-GLOBAL

  • 森林環境モニタリングロボットにおけるSLAMと深層学習を用いた領域推定に基づく経路計画法の検討

    森本純矢, 林拓哉, 藤本裕之, 石井裕之, 大谷淳, 大和淳司, 高西淳夫

    画像電子学会年次大会予稿集(CD-ROM)   46th   ROMBUNNO.S1‐3  2018.06

    J-GLOBAL

  • 環境モニタリングロボットの視覚情報処理を用いた森林における起点への帰還法の検討

    藤本裕之, 石井裕之, 大谷淳, 大和淳司, 高西淳夫

    画像電子学会年次大会予稿集(CD-ROM)   45th   ROMBUNNO.S‐2  2017.06

    J-GLOBAL

  • 縫合手術を構成する6つの段階の動画像中の手術者の手の動きに基づき予測する方法 (ITS)

    坪田 実和, 李 イエ, 大谷 淳

    電子情報通信学会技術研究報告 = IEICE technical report : 信学技報   116 ( 463 ) 257 - 262  2017.02

    CiNii

  • H-012 Study of Ultrasound Image Segmentation Based on SuperVoxel for Extracting Fetal Tissues

    Wang Yuanchen, Tatematsu Naotomo, Ohya Jun, Xu Rong, Ishikawa Makiko, Iwata Hiroyasu

      14 ( 3 ) 149 - 150  2015.08

    CiNii

  • H-008 Study of SLAM Based Confirmation of Results of Mechanical Operations by a Clean-Up Robot

    GAO Si, TATEMATSU Naotomo, OHYA Jun

      14 ( 3 ) 135 - 136  2015.08

    CiNii

  • H-030 A Method for Tracking Guitar Neck and Fingertips: Necking Tracking Robust against Occlusions Based on Geometry Analysis and Fingertips Tracking Based on Temporal Probability Map

    Zhao Wang, Ohya Jun

      14 ( 3 ) 195 - 196  2015.08

    CiNii

  • I-015 Study of an Automatic System that Generates Portraits from Facial Images Using Neural Networks : Comparative Study Based on Subjective Tests

    GUO Shuai, ENOMOTO Makoto, KAWAMURA Harumi, OHYA Jun

      14 ( 3 ) 231 - 232  2015.08

    CiNii

  • I-009 Study of Tracking the Entire Tree from a Video Sequence Based on Tracking a Small Number of Feature Points

    Shen Yuanchao, Ohya Jun

      14 ( 3 ) 217 - 218  2015.08

    CiNii

  • Study of Early Event Recognition Based on 3D Positions of Joints of Human Bodies : Utilizing Skeletons Based on Relative Positions of Joints of Human Bodies in a Lie Group and Max-Margin Early Event Detectors

    TATEMATSU Naotomo, OHYA Jun, DAVIS Larry S.

    Technical Committee on Biometrics   114 ( 520 ) 229 - 234  2015.03

     View Summary

    This paper proposes a method for detecting early events by utilizing 3D Positions of Joints of Human Bodies in RGB and Depth (RGBD) image sequences acquired by Kinect[○!C]. Our method incorporates features based on relative positions of joints of human bodies in a Lie Group into the max-margin early event detectors. Experiments using multiple human actions in real color and depth image sequences demonstrate the effectiveness of our method.

    CiNii

  • D-16-12 Fundamental Study of Early Recognition of Surgeon's Action during Suture Surgery from Video Sequences

    Li Ye, Ohya Jun, Chiba Toshio, Xu Rong, Yamashita Hiromasa

    Proceedings of the IEICE General Conference   2015 ( 2 ) 152 - 152  2015.02

    CiNii

  • D-11-38 Fundamental Study of Tracking Fingers in RGBD Dynamic Image Sequences towards Automatic Teaching System for Piano Practice

    Qiu Dong, Ohya Jun

    Proceedings of the IEICE General Conference   2015 ( 2 ) 38 - 38  2015.02

    CiNii

  • D-12-53 Study of Early Event Recognition by Utilizing Skeletons Based on Relative Positions of Joints of Human Bodies in a Lie Group and Max-Margin Early Event Detectors

    Tatematsu Naotomo, Ohya Jun, Davis Larry S.

    Proceedings of the IEICE General Conference   2015 ( 2 ) 105 - 105  2015.02

    CiNii

  • D-11-12 Study of a Vision Based Method for Checking the Position of Each Finger of Guitar Players : Towards the Actualization of an Autonomous Guitar Chord Teaching System

    WANG Zhao, OHYA Jun

    Proceedings of the IEICE General Conference   2015 ( 2 ) 12 - 12  2015.02

    CiNii

  • 抽象絵画における鑑賞者の視線行動を利用した音楽生成システムの研究-鑑賞者の発話内容の重複と発想の広がりとの関連に関する検討-

    小楠竜也, 大谷淳, 楜沢順, 米村俊一

    電子情報通信学会大会講演論文集(CD-ROM)   2015  2015

    J-GLOBAL

  • 抽象絵画における鑑賞者の注視情報を利用した音楽生成システム-抽象絵画の特徴が鑑賞者の抽象的な発想の変化に及ぼす影響-

    小楠竜也, 大谷淳, 楜沢順, 米村俊一

    情報科学技術フォーラム講演論文集   14th  2015

    J-GLOBAL

  • Study of Recognizing Customers' Behaviors in Shops by Dynamic Image Processing and Bayesian Network

    YAN Jing, HORI Takayuki, OHYA Jun

    Technical report of IEICE. PRMU   114 ( 197 ) 101 - 106  2014.09

     View Summary

    This paper explores the validity of embedding the image processing based extraction of 10 image features that could describe customers' shopping behaviors to a Bayesian Network based classification method that can recognize the four categories: combinations of with/without hesitation and bought/not bought. By constructing a system that consists of the three modules for tracking customers, extracting the image features and classifying the customers' behaviors, recognition tests using 48 video data were conducted. As a result of the experiments, a recognition accuracy of higher than 89% is achieved, which is promising.

    CiNii

  • Study of Recognizing Customers' Behaviors in Shops by Dynamic Image Processing and Bayesian Network

    Jing Yan, Takayuki Hori, Jun Ohya

    IPSJ SIG Notes. CVIM   2014 ( 17 ) 1 - 6  2014.08

     View Summary

    This paper explores the validity of embedding the image processing based extraction of 10 image features that could describe customers' shopping behaviors to a Bayesian Network based classification method that can recognize the four categories: combinations of with/without hesitation and bought/not bought. By constructing a system that consists of the three modules for tracking customers, extracting the image features and classifying the customers' behaviors, recognition tests using 48 video data were conducted. As a result of the experiments, a recognition accuracy of higher than 89% is achieved, which is promising.

    CiNii

  • I-036 Study of Subjective Discrimination in an Automatic System for Generating Line Drawing Based Portraits from Facial Images

    Wu Yuzhen, Enomoto Makoto, Ohya Jun

      13 ( 3 ) 247 - 248  2014.08

    CiNii

  • H-027 Study of Image Features for a Bayesian Network Based Method for Recognizing Customers' Behaviors in Shops

    Yan Jing, Hori Takayuki, Ohya Jun

      13 ( 3 ) 121 - 122  2014.08

    CiNii

  • I-035 Fundamental Study of Virtual Make-up System : Study of Retrieving Made-up Faces Based on Similarities of Face Shapes

    Zhang Lin, Ohya Jun

      13 ( 3 ) 245 - 246  2014.08

    CiNii

  • Study of Portrait Similarities in the Automatic System that Generates Portraits from Facial Images

    Wu Yuzhen, Enomoto Makoto, Ohya Jun

    IEICE technical report. Image engineering   114 ( 172 ) 1 - 6  2014.08

     View Summary

    This paper studies an automatic method that generates portraits by utilizing a neural network trained by facial features obtained by applying Active Shape Model to original face images and by portraits drawn by an illustrator. The authors' conventional research works had remaining issues such as low subjective similarities of generated portraits, validity of the nose, and absence of subjective evaluation of generated portraits. To increase the subjective similarity, this paper proposes a new portrait depicting tool. We obtain promising feeling that subjective similarity can be improved by the new depicting tool. In addition, we conduct experiments on subjective discrimination of portraits. It turns out that the new tool gives higher subjective discrimination rate than our conventional tool, and that the case of the new tool and presence of the nose gives the highest discrimination rate.

    CiNii

  • Study of Clustering Feature points of Moving Objects From Dynamic RGB-D Images : Proposing a Method that Combines Supervoxel and Multi Label Graph-Cut

    Tatematsu Naotomo, Ohya Jun, Davis Larry S.

    Technical report of IEICE. PRMU   113 ( 493 ) 127 - 132  2014.03

     View Summary

    This paper proposes a method that can discriminate feature points of each moving object from the still background by utilizing supervoxel and Multi-label Graph-Cut. The ordinal clustering method based on multi label graph-cut has a problem that can not classify small movement object. To solve this problem, our method utilize initial cluster based on supervoxel and construct multi-labeled-graph by the unit of group that belongs same supervoxel. This improvement enable accurate clustering feature points even if the amount of movement is small. Experiments using multiple moving objects and real stereo sequences demonstrate the effectiveness of our proposed method.

    CiNii

  • Study of Recognizing Hand Actions from Video Sequences during Suture Surgeries Based on Temporally-Sectioned SIFT and Sliding Window Based Neural Networks

    Li Ye, Ohya Jun, Chiba Toshio, Xu Rong, Yamashita Hiromasa

    Technical report of IEICE. PRMU   113 ( 493 ) 151 - 156  2014.03

     View Summary

    Towards the realization of a robotic nurse that can support surgeries autonomously by recognizing surgical situations only using video informations, this paper proposes an improved method by using sectioned-SIFT and sliding window based neural network that can recognize surgeon's hand actions: suture and tying. Hand area is detected by using color information and then the video sequence is partitioned into sections. Sectioned-SIFT descriptors are computed in each section and built a word vocabulary. Histogram feature of the action is spliced by using word's frequency in each section. Finally, sliding window and neural network is used to recognize the significant actions: suture and tying. The proposed method has achieved the 100% recognition rate for manually extracted actions and 90% recognition rate for whole surgery video sequences.

    CiNii

  • Fundamental Study of Recognizing Customers' Behaviors in Shops Based on Facial Expression Analysis : Acquiring Facial Expression Features by ASM and How to Build Bayesian Networks' Models

    JIA Luwei, OHYA Jun, HORI Takayuki

    Technical report of IEICE. PRMU   113 ( 493 ) 157 - 162  2014.03

     View Summary

    Towards the realization of a method that can recognize the psychological state that causes the purchasing behaviors of consumers, this paper proposes a method for recognizing customers' behaviors based on analyses of the customers' facial expressions. First, in each frame of the video sequence, facial feature points are extracted by ASM (Active Shape Model), and features are obtained from feature points in the eye-brows and mouth. Then, each feature is assigned to a node of Bayesian Network. On recognition, by inputting a video sequence, the Bayesian Network outputs the behavior recognition result. The validity of the proposed method is presented experimentally.

    CiNii

  • Study of Tongue Detection and Tongue Shape Classification Based on Active Contour Model and Texture Features

    TIAN Ming, LUO Dan, OHYA Jun

    Technical report of IEICE. PRMU   113 ( 493 ) 133 - 138  2014.03

     View Summary

    In this paper, we aim to an automatic tongue diagnosis system in traditional Chinese medicine. First, we propose a new method to extract the tongue area automatically from digital photos using the active contour model based on the watershed segmentation. Second, we classify the following symptoms: the teeth-printed tongue, which could be diagnosed by the contour of tongue, the tongue coating, the fissured tongue, which cloud be diagnosed by the texture feature of tongue, as well as the swelling tongue or thin tongue, which could be diagnosed by the area of tongue. Finally, we confirm the effectiveness of the proposed method by experiments.

    CiNii

  • Study of the Optimal Number and Locations of Facial Feature Points in Generating Portraits Using Neural Networks

    FANG Guoyong, ENOMOTO Makoto, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   113 ( 470 ) 145 - 150  2014.03

     View Summary

    We study a system that can automatically generate portraits from face images using multi-layered neural networks, for which image features of the face detected by ASM are the inputs, the portrait drawn by an illustrator are teaching signals, and back-propagation algorithms are performed for learning. This system aims at outputting portraits that do not depend on deformations of the line drawing traced from the face image, and captures the drawing style of an individual artist. The number and locations of feature points detected by ASM could affect quality of generated portraits. This paper reports new results that was not clear in our laboratory conventional research. We study the optimal number and locations of the detected facial feature points in generating portraits.

    CiNii

  • D-12-76 Fundamental Study of Recognizing Customers' Entire Bodies' Behaviors in Shops Using Image Processing and Bayesian Network

    Yan Jing, Hori Takayuki, Ohya Jun

    Proceedings of the IEICE General Conference   2014 ( 2 ) 151 - 151  2014.03

    CiNii

  • D-16-5 Study of Recognizing Surgeon's Action during Suture Surgery by Using Sectioned-SIFT

    Li Ye, Ohya Jun, Chiba Toshio, Xu Rong, Yamashita Hiromasa

    Proceedings of the IEICE General Conference   2014 ( 2 ) 169 - 169  2014.03

    CiNii

  • D-12-68 Study of Lip Reading from Video Sequences by Combining Optical Flow Features and Shape Features with Discrete Cosine Transform Features

    Takahashi Shohei, Ohya Jun

    Proceedings of the IEICE General Conference   2014 ( 2 ) 143 - 143  2014.03

    CiNii

  • D-12-45 Study of Tongue Detection and Tongue Shape Classification for Tongue Diagnosis

    Tian Ming, Luo Dan, Ohya Jun

    Proceedings of the IEICE General Conference   2014 ( 2 ) 120 - 120  2014.03

    CiNii

  • D-12-65 Study of Clustering Moving Objects's Feature points by Supervoxel and Multi-label Graph-Cut

    Tatematsu Naotomo, Ohya Jun, Davis Larry S.

    Proceedings of the IEICE General Conference   2014 ( 2 ) 140 - 140  2014.03

    CiNii

  • D-12-75 Study of Recognizing Customers' Behaviors Based on Facial Expression Analysis and Bayesian Networks

    Jia Luwei, Ohya Jun, Hori Takayuki

    Proceedings of the IEICE General Conference   2014 ( 2 ) 150 - 150  2014.03

    CiNii

  • D-11-5 Study of an Automatic Line Drawing Based Portrait Generation System Based on Neural Network : Study of Generating Noses

    Wu Yuzhen, Enomoto Makoto, Ohya Jun

    Proceedings of the IEICE General Conference   2014 ( 2 ) 5 - 5  2014.03

    CiNii

  • D-11-6 Fundamental Study of Generating Portraits Using Neural Networks : Study of the Optimal Number of Facial Feature Points Based on Physical Errors between the Feature Points and Generated Portraits

    Fang Guoyong, Enomoto Makoto, Ohya Jun

    Proceedings of the IEICE General Conference   2014 ( 2 ) 6 - 6  2014.03

    CiNii

  • Automatic Lip-Reading by using Multiple Visual Features -Integration of the Shape, Optical Flaw and DCT features-

    Shohei Takahashi, Jun Ohya

    IPSJ SIG Notes. CVIM   2014 ( 7 ) 1 - 7  2014.02

     View Summary

    In the paper, we present a lip-reading method that can recognize speech by using only visual features. Lip-reading can work well in noisy places such as in the car or in the train. In addition people with hearing-impaired or difficulties in hearing can be benefited. First, the Active Shape Model (ASM) is applied to track and detect the face and lip in a video sequence. Second, three visual features, the shape, optical flow and Discreet cosine transformation of the lip are obtained from the lip area detected by ASM. The extracted features are ordered chronologically so that Support Vector Machine (SVM) is performed so as to learn and classify the spoken words. Hierarchical SVMs are used to recognize the words. Each visual feature is trained by the respective middle-layer SVM, and those outputs of SVM's are integrated by the final SVM. Experimental results show that the integration of these features improves the recognition accuracy.

    CiNii

  • 抽象絵画の鑑賞者の視線行動に基づく音楽生成システムの予備的検討-音楽の有無が抽象絵画鑑賞者の発想の多寡に与える影響-

    小楠竜也, 大谷淳, 楜沢順, 米村俊一

    電子情報通信学会大会講演論文集(CD-ROM)   2014  2014

    J-GLOBAL

  • Fundamental Study of a Music Generation System Based on Gaze Behaviors of Viewers of Abstract Paintings-Effects of Proposed System on the Amount of Ideas of Viewers-

    小楠竜也, 大谷淳, 楜沢順, 米村俊一

    電子情報通信学会技術研究報告   113 ( 501(HIP2013 78-105) )  2014

    J-GLOBAL

  • 抽象絵画における鑑賞者の視線行動を利用した音楽生成システムの研究-鑑賞者の発想を膨らませる音楽の自動生成システムに関する予備的検討-

    小楠竜也, 大谷淳, 楜沢順, 米村俊一

    情報科学技術フォーラム講演論文集   13th  2014

    J-GLOBAL

  • Study of Structure-from-motion Based 3D Reconstruction and Surface Mosaic

    Duan Mengyao, Xu Rong, Ohya Jun

      113 ( 197 ) 113 - 118  2013.09

     View Summary

    Structure-from-motion (SFM) is a significant issue of 3D reconstruction for many applications. However, some image collector like endoscopy only provides a narrow view angle and 2D structures of the internal organs. In order to expand the view angle and describe the details of 3D structures, we propose an approach to reconstruct 3D surface based on image stitching from a video sequence. There are four major steps of the proposed method: 3D reconstruction, 3D point cloud stitching, dense point cloud generation and Poisson surface reconstruction. Experimental results demonstrate that the proposed method achieves a satisfied accuracy and efficiency for 3D reconstruction, and promise a large view angle.

    CiNii

  • Study of Recognizing Spontaneous Facial Expressions of a Person who Watches Web News Based on ASM and Bayesian Network

    XU Chao, OHYA Jun

      113 ( 197 ) 7 - 12  2013.09

     View Summary

    To achieve a comment system that can automatically estimate emotions of a person who watches web news, this paper proposes a computer vision based method that can recognize facial expressions generated in response to Web news. During the learning session, first, ASM (Active Shape Model) is applied to the face so that some feature points are located. From the located feature points, some features such as distances between feature points are obtained. The features collected from sample data are used for nodes of Bayesian Networks. During facial expression estimation session, the features that are computed from the input facial video sequence by the same procedure as the learning session are input to the Bayesian Networks so that the recognition result is obtained as the facial expression category that maximizes the a-posteriori probability. Experimental results demonstrate the effectiveness of the proposed method.

    CiNii

  • Study of Recognizing Spontaneous Facial Expressions of a Person who Watches Web News Based on ASM and Bayesian Network

    XU Chao, OHYA Jun

    Technical report of IEICE. PRMU   113 ( 196 ) 7 - 12  2013.09

     View Summary

    To achieve a comment system that can automatically estimate emotions of a person who watches web news, this paper proposes a computer vision based method that can recognize facial expressions generated in response to Web news. During the learning session, first, ASM (Active Shape Model) is applied to the face so that some feature points are located. From the located feature points, some features such as distances between feature points are obtained. The features collected from sample data are used for nodes of Bayesian Networks. During facial expression estimation session, the features that are computed from the input facial video sequence by the same procedure as the learning session are input to the Bayesian Networks so that the recognition result is obtained as the facial expression category that maximizes the a-posteriori probability. Experimental results demonstrate the effectiveness of the proposed method.

    CiNii

  • Study of Structure-from-motion Based 3D Reconstruction and Surface Mosaic

    Duan Mengyao, Xu Rong, Ohya Jun

    Technical report of IEICE. PRMU   113 ( 196 ) 113 - 118  2013.09

     View Summary

    Structure-from-motion (SFM) is a significant issue of 3D reconstruction for many applications. However, some image collector like endoscopy only provides a narrow view angle and 2D structures of the internal organs. In order to expand the view angle and describe the details of 3D structures, we propose an approach to reconstruct 3D surface based on image stitching from a video sequence. There are four major steps of the proposed method: 3D reconstruction, 3D point cloud stitching, dense point cloud generation and Poisson surface reconstruction. Experimental results demonstrate that the proposed method achieves a satisfied accuracy and efficiency for 3D reconstruction, and promise a large view angle.

    CiNii

  • I-046 Fundamental Study of Generating Portraits Using Neural Networks : Study of Relationship between Face Features Detected by ASM and the Generated Portrait Quality

    Fang Guoyong, Enomoto Makoto, Ohya Jun

      12 ( 3 ) 347 - 348  2013.08

    CiNii

  • H-034 Fundamental Study of Recognizing Customers' Entire Bodies' Behaviors in Shops Using Image Processing and Bayesian Network

    Yan Jing, Hori Takayuki, Ohya Jun

      12 ( 3 ) 169 - 170  2013.08

    CiNii

  • H-013 Study of Extracting Moving Objects from Dynamic Kinect Images and Segmenting the Moving and Still Objects' Reconstructed 3D Shape : Proposing a combined method Probabilistic Boundary Edge Map with supervoxel

    Tatematsu Naotomo, Ohya Jun, Davis Larry

      12 ( 3 ) 119 - 120  2013.08

    CiNii

  • H-008 Improving a Bayesian Network Based Recognition of Spontaneous Facial Expressions of a Person who Watches Web News : Utilizing Image Features for Blinks

    Xu Chao, Ohya Jun

      12 ( 3 ) 109 - 110  2013.08

     View Summary

    Recently reading news through web news media is becoming popular. Most of web news is delivered together with a comment system, which asks the user to rate items such as "the news let you think", and "boring". However, many users do not like manual operations for rating; therefore, not many users actually input the rates. One solution for this issue is to utilize results of recognizing facial expressions from the video sequence acquired by the camera that observes the user. So far, there are very many works on recognizing six fundamental expressions such as sad, surprise and happy, for example, by HMM (Hidden Markov Models). However, not many works dealt with recognizing expressions that could appear when the user watches web news; main difficulty in recognizing these expressions could be caused by the fact that these expressions are spontaneously generated. It is difficult to recognize the spontaneous expressions, which come with tiny movements hard to be detected. This paper proposes a method that aims at recognizing spontaneous expressions.

    CiNii

  • H-007 Visual-based Human Gesture Recognition Using Multi-modal Features

    Luo Dan, Ohya Jun

      12 ( 3 ) 107 - 108  2013.08

    CiNii

  • H-038 Fundamental Study of Recognizing the Surgeon's Action during Suture Surgery from the Video Sequence

    Li Ye, Ohya Jun, Chiba Toshio, Xu Rong, Yamashita Hiromasa

      12 ( 3 ) 179 - 180  2013.08

    CiNii

  • H-012 Study of tongue detection and feature extraction for Tongue Diagnosis

    Tian Ming, Luo Dan, Ohya Jun

      12 ( 3 ) 117 - 118  2013.08

    CiNii

  • H-019 Fundamental Study of Particle Filter based Motorbike Tracking for a Violator Detection System

    Do Khanh N., Ohya Jun

      12 ( 3 ) 135 - 136  2013.08

    CiNii

  • H-047 Endoscope based gastric 3D reconstruction and surface mosaic

    Duan Mengyao, Xu Rong, Ohya Jun

      12 ( 3 ) 197 - 198  2013.08

    CiNii

  • K-055 Fundamental Study of Lip-Reading Using Record of Optical Flow Directions and Support Vector Machine

    Takahashi Shohei, Ohya Jun

      12 ( 3 ) 681 - 682  2013.08

    CiNii

  • E-024 Research of Automatic Composition Based on Viewer's Gaze Behavior on Abstract Painting : Fundamental Study of a Method of Generating Tonal Music from the Pattern of Gaze Behavior

    Ogusu Tatsuya, Ohya Jun, Kurumisawa Jun

      12 ( 2 ) 237 - 238  2013.08

    CiNii

  • Study of Robust Facial Feature Localization by Modified Active Shape Model

    Li Yan, Luo Dan, Jun Ohya

    Technical report of IEICE. PRMU   112 ( 495 ) 129 - 134  2013.03

     View Summary

    This paper proposes a modified Active Shape Model (ASM), which is featured by the following two modules: (1)A flexible method that automatically generates training set and does not require time-consuming manual operations, which are required by the conventional ASM; (2)Two-step fitting that consists of the initial global (rough)fitting for the entire face and the subsequent local (precise)fitting for each facial component, where both fittings are based on ASM. Experiments demonstrate that more accurate location results were obtained by the proposed method than the conventional ASM.

    CiNii

  • Fundamental Study of Estimating the Light Source Position by Comparing the 3D Reconstruction Obtained from RGB Image with Depth Image

    Hayashida Tomoki, Takahashi Shohei, Okuya Takeshi, Ohya Jun, Kawamura Harumi

    Technical report of IEICE. Multimedia and virtual environment   112 ( 474 ) 145 - 150  2013.03

     View Summary

    This paper proposes a method for estimating the light source direction from RGB image and Depth image (K-D image), which can be acquired by Kinect sensor etc. The proposed method utilizes the property that the 3D reconstructed result, which is obtained by converting the RGB image to gray-level image and then converting that gray-level image to depth image (D-D image) by Khan et al. 's method based on "Deep is dark paradigm", is deformed to the light source direction. The value difference at each pixel of the D-D image and K-D image is obtained, and that pixel's surface normal is computed from the K-D image. In the 2D space defined by the two angle parameters that represent the surface normal's direction, each pixel's difiFerence is added to the cell that corresponds to that pixel's normal's direction. After this process is performed for all the pixels, the light source direction is determined based on the peak in the parameter space. Experiments using some diflferent objects under different light source directions were performed. Promising results were obtained.

    CiNii

  • Research on Recognizing Customers' Behaviors in Shops Based on Bayesian Network

    ZHENG Yao, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   112 ( 474 ) 51 - 56  2013.03

     View Summary

    For supermarkets and stores, marketing surveillance is a very important issue: in particular, customers' hesitations on purchase, which cannot be recorded by POS systems, are extremely important. This paper studies how to classify the following four categories: "purchased without hesitation", "purchased with hesitation", "not purchased without hesitation", and "not purchased with hesitation". Since these behaviors tend to vary depending on persons and situations, Bayesian network is considered to be a useful method for the classification. By manually extracting image features from video sequences that observe a shop's comer, Bayesian network based classification was explored. Consequently, promising experimental results were obtained.

    CiNii

  • Fundamental Study of Judging Sleeper's Turnovers from Depth Image Sequences by Support Vector Machine

    Minezaki Yuta, Ohya Jun

    Technical report of IEICE. Multimedia and virtual environment   112 ( 474 ) 107 - 112  2013.03

     View Summary

    In recent years, care of aged people's turnovers is heavy burdens for helpers; in particular, automatic judgment of turnovers is desired. This paper proposes a method that can judge whether aged people being cared turned over by utilizing a depth image sequence acquired by Kinect sensor that observes the aged people in beds. Specifically, image features are extracted from aged people's bodies in the depth image sequences. and are used for classification by Support Vector Machine. As the image features, this paper uses time series data of "xyz differences between left and right shoulders and between the left and right edges of the waist". and "average depth in each block obtained by partitioning the bed area into w by h blocks" for exploring classification accuracies. Experimental results show that the former and latter features achieve 99.5% and 89.5% classification accuracies, respectively. The fonner gives a better accuracy, but its applicability is limited. while the latter can be applied for more general cases.

    CiNii

  • Research on Detecting Strangers from Multiple Pan-Tilt-Zoom Static Cameras

    MIZOGUCHI Yoshinori, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   112 ( 474 ) 45 - 50  2013.03

     View Summary

    In this paper, we propose an approach that captures a distant person's face image with high resolution by zooming the camera. Like an ordinary security system, cameras observe down from upper positions. The proposed approach extracts human skin color areas and zooms to the estimated position of the face, even if the resolution of the initial face image is so low that the face cannot be recognized. The captured face image is converted to a DCT (Discrete Cosine Transform)based feature vector, and then SVM (Support Vector Machine) judges if the person is a stranger. The proposed approach shares image information obtained by multiple cameras, and limits the number of zooming camera to one based on 3D position of the moving human in the continuous frames. This could achieve that both large area surveillance and face recognition can be performed simultaneously. Experimental results show validity of the proposed approach.

    CiNii

  • D-12-76 Research on Detecting Strangers from Multiple Pan-Tilt-Zoom Static Cameras

    Mizoguchi Yoshinori, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 169 - 169  2013.03

    CiNii

  • D-16-5 A Robust ICP Algorithm for Automatic Registration of 3D Fetal Facial Surface

    Xu Rong, Ohya Jun, Zhang Bo, Sato Yoshinobu, Fujie Masakatsu

    Proceedings of the IEICE General Conference   2013 ( 2 ) 196 - 196  2013.03

    CiNii

  • D-12-72 Fundamental Study of Judging Sleepers' Turnovers from RGBD ImageSequences by Support Vector Machine

    Minezaki Yuta, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 165 - 165  2013.03

    CiNii

  • D-12-80 Temporal Modified-RANSACとSupervoxelべ一スのグラフカットを用いた移動RBGDカメラからの移動物体領域のセグメンテ-ションの検討

    Tatematsu Naotomo, Ohya Jum, Davis Larry's.

    Proceedings of the IEICE General Conference   2013 ( 2 ) 173 - 173  2013.03

    CiNii

  • D-11-20 Fast Computation of Curvatures for Polygon Models Using Normal Vectors

    Okuya Takeshi, Yabuno Ken, Ohya Jun, Takahashi Nobuyuki

    Proceedings of the IEICE General Conference   2013 ( 2 ) 20 - 20  2013.03

    CiNii

  • D-12-12 Bayesian network model construction for Recognizing Customers, Behaviors in Shops

    ZHENG YAO, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 105 - 105  2013.03

    CiNii

  • D-12-11 Study of Recognizing Spontaneous Facial Behaviors Generated by Persons Watching Web News by Combining ASM and Bayesian Network

    Xu Chao, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 104 - 104  2013.03

    CiNii

  • D-12-10 Face Recognition Using Sparse Representation and Moments

    Li Ye, Shu Huazhong, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 103 - 103  2013.03

    CiNii

  • D-12-21 Robust Lip Contour Localization by Modified Active Shape Model

    Yan Li, Dan Luo, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 114 - 114  2013.03

    CiNii

  • D-11-78 A Study on 3D Point Cloud Stitching for SFM Reconstruction

    Duan Mengyao, Xu Rong, Ohya Jun

    Proceedings of the IEICE General Conference   2013 ( 2 ) 78 - 78  2013.03

    CiNii

  • A Study on Illuminant Color Estimation based on Gray World Assumption using High and Low Chroma Gamuts

    川村春美, 米村俊一, 大谷淳, 小島明

    情報処理学会研究報告(CD-ROM)   2012 ( 6 )  2013

    J-GLOBAL

  • A Study of Automatic 3D Fetal Face Detection By Locating Facial Features From 3D Ultrasound Images for Navigating FETO Surgeries

    XU Rong, OHYA Jun, ZHANG Bo, SATO Yoshinobu, Fujie Masakatsu G.

    IEICE technical report.   112 ( 271 ) 59 - 64  2012.10

     View Summary

    With the wide clinical application of 3D ultrasound (US) imaging, automatic location of fetal facial features from us volumes for navigating fetoscopic tracheal occlusion (FETO) surgeries becomes possible, which plays an important role in reducing surgical risk. In this paper, we propose a feature-based method to automatically detect 3D fetal face and accurately locate key facial features without any priori knowledge or training data. The candidates of the key facial features, such as the nose, eyes, nose upper bridge and upper lip are detected by analyzing the mean and Gaussian curvatures of the facial surface. Each feature is gradually identified from the candidates by a boosted cascade filtering scheme based on the spatial relations between each feature. In experiments, an identification rate of 100% is achieved by using 72 3D US images from a test database of 6 fetal faces in the frontal view and any pose within 15° from the frontal view, and the location error 3.18±0.91mm of the detected upper lip for all test data is obtained, which can be tolerated by the FETO surgery.

    CiNii

  • H-020 Study of Extracting Moving Objects from Dynamic Kinect Images and Segmenting the Moving and Still Objects' Reconstructed 3D Shape : Proposing a Fixation-based Segmentation Method Utilizing Probabilistic Boundary Edge Map

    Tatematsu Naotomo, Ohya Jun, Davis Larry

      11 ( 3 ) 161 - 162  2012.09

    CiNii

  • H-046 Fundamental Study of Recognizing Customers' Behaviors in Shops Based on Bayesian Network

    YAO ZHENG, Ohya Jun

      11 ( 3 ) 221 - 222  2012.09

    CiNii

  • H-048 Fundamental Study of Reconstructing Botanical Tree's 3D Models from Kinect's Images and Dynamical Behaviors of Non-leafy Trees

    Do Khanh N., Ohya Jun

      11 ( 3 ) 225 - 226  2012.09

    CiNii

  • H-042 Fundamental Study of Detecting Strangers from Surveillance Camera

    Mizoguchi Yoshinori, Ohya Jun

      11 ( 3 ) 213 - 214  2012.09

    CiNii

  • H-047 A Study of Extracting 3D Facial Feature from Kinect's Image by Integrating ASM and Depth Map

    LI YAN, LUO DAN, OHYA JUN

      11 ( 3 ) 223 - 224  2012.09

    CiNii

  • H-045 Study of Extracting Multi-modal features for Recognizing Sign Language Vocabularies that Combine Facial Expressions and Hand Gestures

    Luo Dan, Ohya Jun

      11 ( 3 ) 219 - 220  2012.09

    CiNii

  • G-014 A Registration Between 3D Ultrasound Images and 3D Fetal Model for Locating a Fetal Mouth in a Fetal Surgical Navigation System

    Xu Rong, Ohya Jun, Sato Yoshinobu, Fujie Masakatsu G.

      11 ( 2 ) 373 - 374  2012.09

    CiNii

  • Object Tracking Using Kinect and its Application in Tracking Dynamical Behaviors of Botanical Tree Leaves

    DO Khanh N., OHYA Jun

    Technical report of IEICE. HIP   111 ( 500 ) 159 - 164  2012.03

     View Summary

    Reconstructing the dynamical behavior of a moving non-rigid object such as smoke, water and botanical tree is a tough task because it is very difficult to track the behavior of such objects due to their complicated structure and behavior. This paper studies on tracking the dynamical behaviors of a botanical tree by particle filter based methods. The proposed method uses particle filter, which is also known as Sequential Monte Carlo method, to estimate the position of the target leaf (leaf to be tracked) in each frame of the video sequences. The motion model predicts the position of the target leaf by generating different hypothesis of the target leaf's positions. The observation model corrects the prediction by evaluating each "particle" based on template matching score between the leaf model (template) and the target leaf. To increase the tracking accuracy and decrease the computation cost, a leaf is only searched for in its region of interest (ROI). The ROI position and size are updated by one frame delay method.

    CiNii

  • A Cone Based Electromagnetic 3D Ultrasound Calibration

    XU Rong, OHYA Jun, ZHANG Bo, SATO Yoshinobu, FUJIE Masakatsu G.

    Technical report of IEICE. HIP   111 ( 500 ) 253 - 258  2012.03

     View Summary

    For a surgical navigation system with swept-volume 3D ultrasound images, it is necessary to estimate the extracted positions of ultrasound voxels relative to a tracking reference by a 3D ultrasoud calibration. In this paper, we propose a cone based 3D ultrasound calibration system by using an electromagnetic tracking device. The proposed phantom is not expensive, easy to construct, easy to scan, and has a simple experiment setup, while the identification of the tips on each cone from a single ultrasound volume is a manual process, the tracking position of each tip is measured by a pen probe, and the calibration matrix is estimated by SVD (Singular Value Decomposition) method. The precision of the system is experimentally evaluated. It turns out that the maximum distance root mean square (RMS) error is 3.32 mm by using 12 cones, which satisfies the requirement of our fetal surgical navigation system: less than 3.85 mm.

    CiNii

  • Analyzing the "●" Communication System's Effect on Users' Self-Disclosures during Tele-Communication Session

    CHEN Li Jen, OHYA Jun, YONEMURA Shunichi, TOKUNAGA Yukio

    Technical report of IEICE. HIP   111 ( 500 ) 123 - 128  2012.03

     View Summary

    This paper analyzes how our Prompter Communication System (PCS), which has single or several black circular images "●" embedded in the background of the text-field canvas to be used for tele-communication between two users, affects telel-communication users' self-disclosures. To the analysis, this research applies the concepts and features of psychological projection techniques and tests. According to these concepts and features, PCS is expected to enhance and encourage the users' self-disclosure, which is essential to developing intimate and deep interpersonal relationship. Communication experiments by using the prototype system, PCS, are conducted, where a Normal Communication System (NCS), which uses only plain text-field canvas without black circles, is compared with PCS. Two subjects are required to make tele-communication using PCS or NCS. PCS/NCS user writes texts to the canvas with/without randomly generated black circles and sends it (message) to the other user. After tele-communication experiments using PCS and NCS, to each message, for coders give a self-disclosure level, which modifies Daibo et al.'s self-disclosure levels. Consequently, it turns out that PCS gives significantly higher frequencies for the high self-disclosure level, which corresponds to the users' inner thoughts and emotions, than NCS.

    CiNii

  • A Cone Based Electromagnetic 3D Ultrasound Calibration

    XU Rong, OHYA Jun, ZHANG Bo, SATO Yoshinobu, FUJIE Masakatsu G.

    Technical report of IEICE. PRMU   111 ( 499 ) 253 - 258  2012.03

     View Summary

    For a surgical navigation system with swept-volume 3D ultrasound images, it is necessary to estimate the extracted positions of ultrasound voxels relative to a tracking reference by a 3D ultrasoud calibration. In this paper, we propose a cone based 3D ultrasound calibration system by using an electromagnetic tracking device. The proposed phantom is not expensive, easy to construct, easy to scan, and has a simple experiment setup, while the identification of the tips on each cone from a single ultrasound volume is a manual process, the tracking position of each tip is measured by a pen probe, and the calibration matrix is estimated by SVD (Singular Value Decomposition) method. The precision of the system is experimentally evaluated. It turns out that the maximum distance root mean square (RMS) error is 3.32 mm by using 12 cones, which satisfies the requirement of our fetal surgical navigation system: less than 3.85 mm.

    CiNii

  • Object Tracking Using Kinect and its Application in Tracking Dynamical Behaviors of Botanical Tree Leaves

    DO Khanh N., OHYA Jun

    Technical report of IEICE. PRMU   111 ( 499 ) 159 - 164  2012.03

     View Summary

    Reconstructing the dynamical behavior of a moving non-rigid object such as smoke, water and botanical tree is a tough task because it is very difficult to track the behavior of such objects due to their complicated structure and behavior. This paper studies on tracking the dynamical behaviors of a botanical tree by particle filter based methods. The proposed method uses particle filter, which is also known as Sequential Monte Carlo method, to estimate the position of the target leaf (leaf to be tracked) in each frame of the video sequences. The motion model predicts the position of the target leaf by generating different hypothesis of the target leaf's positions. The observation model corrects the prediction by evaluating each "particle" based on template matching score between the leaf model (template) and the target leaf. To increase the tracking accuracy and decrease the computation cost, a leaf is only searched for in its region of interest (ROI). The ROI position and size are updated by one frame delay method.

    CiNii

  • Analyzing the "●" Communication System's Effect on Users' Self-Disclosures during Tele-Communication Session

    CHEN Li Jen, OHYA Jun, YONEMURA Shunichi, TOKUNAGA Yukio

    Technical report of IEICE. PRMU   111 ( 499 ) 123 - 128  2012.03

     View Summary

    This paper analyzes how our Prompter Communication System (PCS), which has single or several black circular images "●" embedded in the background of the text-field canvas to be used for tele-communication between two users, affects telel-communication users' self-disclosures. To the analysis, this research applies the concepts and features of psychological projection techniques and tests. According to these concepts and features, PCS is expected to enhance and encourage the users' self-disclosure, which is essential to developing intimate and deep interpersonal relationship. Communication experiments by using the prototype system, PCS, are conducted, where a Normal Communication System (NCS), which uses only plain text-field canvas without black circles, is compared with PCS. Two subjects are required to make tele-communication using PCS or NCS. PCS/NCS user writes texts to the canvas with/without randomly generated black circles and sends it (message) to the other user. After tele-communication experiments using PCS and NCS, to each message, for coders give a self-disclosure level, which modifies Daibo et al.'s self-disclosure levels. Consequently, it turns out that PCS gives significantly higher frequencies for the high self-disclosure level, which corresponds to the users' inner thoughts and emotions, than NCS.

    CiNii

  • Research on Tracking Fetus' Mouth in Endoscope Video Sequence for Supporting Fetal Surgeries

    XIE Tianliang, XU Rong, OHYA Jun, ZHANG Bo, SATO Yoshinobu, FUJIE Masakatsu G.

    Technical report of IEICE. PRMU   111 ( 499 ) 235 - 240  2012.03

     View Summary

    An ordinary treatment for Congential Diaphragmatic Hernia is Fetal Endoscopic Tracheal Occlusion (FTO), by which a balloon is placed in the trachea so that the lung can be restored in a natural fashion. However, this treatment required very dangerous surgery; thereby, a supporting system for the surgery is demanded. In this paper, to achieve a system that can automatically navigate a tool to be inserted into the fetal trachea, a method that can detect the mouth of the fetus from the endoscopic video sequence is explored. More specifically, a method that combines ellipse detection by randomized Hough transform with CamShift is proposed. Experimentally, the validity of the proposed method is confirmed.

    CiNii

  • A-15-6 Hierarchical Structure for Memories of Faces in Analytic Representation Area : Application to Computer-based Caricatured Portrait Generation System

    Enomoto Makoto, Ohya Jun

    Proceedings of the IEICE General Conference   2012   226 - 226  2012.03

    CiNii

  • A-15-8 Fundamental Study of Generating Portraits Using Neural Networks

    TAKANO Akira, OHYA Jun, ENOMOTO Makoto

    Proceedings of the IEICE General Conference   2012   228 - 228  2012.03

    CiNii

  • A-16-5 Analysis of Gaze Behavior of Viewers Who See Abstract Paintings and Proposal of its Application to Music Composition

    Ogusu Tatsuya, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2012   254 - 254  2012.03

    CiNii

  • D-12-15 Human gesture recognition using multimodal features

    Luo Dan, Gao Hua, Ekenel Hazim Kemal, Ohya Jun

    Proceedings of the IEICE General Conference   2012 ( 2 ) 109 - 109  2012.03

    CiNii

  • D-12-13 Real Length Measurement by Kinect

    Duan Mengyao, Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2012 ( 2 ) 107 - 107  2012.03

    CiNii

  • D-12-48 Study of the Particle Filter Based Method for Tracking the Dynamical Behaviors of Botanical Tree Leaves in Video Sequence

    Do Khanh N., Ohya Jun

    Proceedings of the IEICE General Conference   2012 ( 2 ) 142 - 142  2012.03

    CiNii

  • D-15-18 Study on a Method of Estimating Learners' Subjective Impressions of the Difficulty for e-Learning on Demand

    Ohkawauchi Takaaki, Ohya Jun, Yonemura Shunichi, Tokunaga Yukio

    Proceedings of the IEICE General Conference   2012 ( 1 ) 215 - 215  2012.03

    CiNii

  • Hierarchical Structure for Memories of Faces in Analytic Representation Area, and Application to Caricatured Portrait Generate System Using Computer

    ENOMOTO Makoto, TAKANO Akira, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   111 ( 479 ) 19 - 24  2012.03

     View Summary

    Viewing the process of drawing caricatured portrait by artist myself introspectively, and it to be a computer program called 'Automatic caricatured portrait generate systems'.

    CiNii

  • Study of Example-based Variations in Virtual Face Makeup Styles

    CHO Jieun, PARK Minje, PARK Sang Il, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   111 ( 479 ) 25 - 29  2012.03

     View Summary

    In this paper, we present a new face makeup analysis and synthesis method for a rapid prototyping of various facial makeup styles. By taking an image pair of a person with and without makeup, we first measure the effect of the given makeup. Then, the extracted makeup is represented by using a finite mixture model. Based on this representation, we provide several operations to easily synthesize a variety of new face makeup effects such as transferring, transplanting, exaggeration, and attenuation. We also demonstrate that multiple makeup styles can be effectively blended to generate a new style in our framework.

    CiNii

  • Fundamental Study of Generating Portraits Using Neural Networks : Studies of Neural Network's Structures and Portraits' Physical and Psychological Similarity

    TAKANO Akira, ENOMOTO Makoto, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   111 ( 479 ) 13 - 18  2012.03

     View Summary

    This paper studies a system that can automatically generate portraits from face images using multi-layered neural networks, for which image features of the face are the inputs, the portrait drawn by an illustrator are teaching signals, and back-propagation algorithms are performed for learning. This system aims at outputting portraits that do not depend on deformations of the line drawing traced from the face image, and reflect human feeling. First, this paper explores the neural networks' structures and relationships between the learning's parameter values and the output results. Second, similarity between the system's outputs and portraits drawn by an illustrator is evaluated by subjective tests, and then, using the evaluation results, the relationship between physical errors (physical quantity) and subjective scores (psychological quantity) is studied.

    CiNii

  • Study of Music Composition Utilizing Gaze Behaviors of Persons Viewing Abstract Paintings

    Tatsuya Ogusu, Jun Ohya, Jun Kurumisawa

    IPSJ SIG Notes   2012 ( 18 ) 1 - 6  2012.01

     View Summary

    This paper proposes a method that allows music composition based on abstract paintings. Our approach utilizes gaze behaviors of viewers who see abstract paintings, where temporal changes in gaze positions are utilized as temporal information for music. In this paper, we measured the viewers' gaze behaviors when they see different abstract paintings, and found the following things: (1) durations in which the fix one's gaze on some position are classified into "average length" group and "extremely long" group, and (2) the number of gaze's stay in each element depends on the element's size and distance from the center of the painting. By analyzing mapping between color and sound based on synesthesia, the element at which the gaze stays is converted to sound so that a new music composition method for generating tone clusters is shown.

    CiNii

  • e-learningにおける主観的難易度の補足方法に対する基礎的検討

    大川内隆朗, 大谷淳, 米村俊一, 米村俊一, 徳永幸生

    電子情報通信学会大会講演論文集   2012  2012

    J-GLOBAL

  • WEB上で動作する“黒丸”プロンプタを用いたコミュニケーションシステム-物語をベースにした複数ユーザ間コミュニケーションの特性分析-

    小駒ちなみ, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    情報処理学会全国大会講演論文集   74th ( 4 )  2012

    J-GLOBAL

  • 低彩度および高彩度色領域を用いた灰色仮説に基づく照明光色推定法の一提案

    川村春美, 米村俊一, 大谷淳, 小島明

    画像電子学会年次大会予稿集(CD-ROM)   40th  2012

    J-GLOBAL

  • e-learningにおける学習者の行動履歴を利用した講義改善支援システムの開発

    大川内隆朗, 大谷淳, 米村俊一, 徳永幸生

    情報科学技術フォーラム講演論文集   11th  2012

    J-GLOBAL

  • Improving Iterative Randomized Hough Transform for Automatic Detection of Fetal Head from Ultrasound Images

    Xu Rong, Ohya Jun, Zhang Bo, Sato Yoshinobu, Fujie Masakatsu G.

    IEICE technical report. Image engineering   111 ( 349 ) 127 - 132  2011.12

     View Summary

    In this paper, we propose an improved iterative randomized Hough transform (IRHT) method to detect fetal head automatically in ultrasound images. With the update of region of interest (ROI) in the IRHT method, the noise pixels are gradually excluded from the region of interest during iteration process, and the estimation becomes progressively close to the target. In order to enhance the efficiency and stability of this algorithm, we consider introducing the number (N) of pixels on the detected ellipses, and select the ellipse with the maximal number of the pixels on the ellipse as the result for each iteration, which is selected from the top-M peaks in the accumulators of the whole detected ellipse samples. The experiments on fetal ultrasound images demonstrate that the proposed method achieves more robust and accurate results, and has a better performance for fetal head detection than the IRHT method.

    CiNii

  • J-036 Fundamental Study of the Analysis of the Gaze Behavior of the Viewer for Abstract Paintings and its Application to Automatic Music Composition

    Ogusu Tatsuya, Ohya Jun, Kurumisawa Jun

      10 ( 3 ) 617 - 618  2011.09

    CiNii

  • H-064 Studies of the Kalman filer and Particle filter method for tracking the moving botanical tree leaves in video sequence

    Khanh Do N., Ohya Jun

      10 ( 3 ) 247 - 248  2011.09

    CiNii

  • H-014 Study of Extracting Moving Objects from Dynamic Stereo Images and Segmenting the Moving and Still Objects' Reconstructed 3D Shape : Proposing a SIFT Feature Based Method Utilizing Energy Function for Foreground or Background

    Tatematsu Naotomo, Ohya Jun

      10 ( 3 ) 133 - 134  2011.09

    CiNii

  • G-004 A Study of Automated Fetal Head Detection by Pre-processing based on Ultrasound Image Gray Feature and Iterative Randomized Hough Transform

    Xu Rong, Ohya Jun, Zhang Bo, Sato Yoshinobu, Fujie Masakatsu G.

      10 ( 2 ) 529 - 530  2011.09

    CiNii

  • Study of Stereoscopic Enhancement for 3D Real Images : Comparative Study of Combined Methods Based on Color Saturation and Brightness

    KU CHENG YU, OHYA Jun, ABE Nobuaki, KAWAI Takashi

    IEICE technical report   110 ( 467 ) 317 - 322  2011.03

     View Summary

    We aim at the histogram equalization method of the weights computed using the local contrast and the distance value at the corresponding pixel in the range image that is a study in the laboratory in the past (WHE), improving the point that only the luminosity of the color image was emphasized. Specifically, using five color contrast enhancement methods to emphasize contrast of the color information except the luminosity and examine a case independent, to use six kinds of methods including WHE. Furthermore, examining five kinds of methods to put WHE together. Therefore, 11 kinds of contrast enhancement methods was totally experimentally compared and found the effective stereoscopic contrast enhancement methods.

    CiNii

  • Study of Extracting Moving Objects from Dynamic Stereo Images and Segmenting the Moving and Still Objects' Reconstructed 3D Shape : Proposing a Method that Combines Temporal Modified-RANSAC and Graph-cut Utilizing Color, Structure and A-priori Probabilities

    TATEMATSU Naotomo, OHYA Jun

    IEICE technical report   110 ( 467 ) 253 - 258  2011.03

     View Summary

    This paper proposes a method that combines Temporal modified RANSAC and Graph-cut utilizing color, structure and A-priori Probabilities. This method can discriminate each moving object from the still background, can compute the stereo cameras' egomotion, and can reconstruct 3D structure of each moving object and static background. Our Temporal modified RANSAC segments the detected 3D optical flows into moving objects and background and computes the rotation and translation matrix for each object. Next, 3d points computed from stereo images segmented into each moving object or static background by Graph-cut. Finally, the 3D points are registered using each object's matrix to the initial positions so that the 3D structures of the moving objects and still background are reconstructed. Experiments using multiple moving objects and real stereo sequences demonstrate the effectiveness of our proposed method.

    CiNii

  • D-12-68 A Study on Edge Feature and its Descriptor

    Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2011 ( 2 ) 171 - 171  2011.02

    CiNii

  • D-12-83 Study of Recognizing Human Motion Observed from an Arbitrary Viewpoint based on Tensor Decomposition Approach

    Hori Takayuki, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2011 ( 2 ) 186 - 186  2011.02

    CiNii

  • D-16-2 A Study of Segmenting Ultrasound Image Based on Bilateral Filtering and Fuzzy Clustering

    Xu Rong, Ohya Jun, Zhang Bo, Sato Yoshinobu, Fujie Masakatsu G.

    Proceedings of the IEICE General Conference   2011 ( 2 ) 205 - 205  2011.02

    CiNii

  • D-12-88 Study of Tracking White Road Line from a Dynamic Video Sequence by a Particle Filter Based Method for Assisting Visually Handicapped People to Walk

    Takahashi Shouhei, Ohya Jun

    Proceedings of the IEICE General Conference   2011 ( 2 ) 191 - 191  2011.02

    CiNii

  • D-12-94 Study of Extracting Moving Objects from Dynamic Stereo Images by Temporal Modified-RANSAC and Graph-Cut and Segmenting the Moving Objects' Reconstructed 3D Shape Data

    Tatematsu Naotomo, Ohya Jun

    Proceedings of the IEICE General Conference   2011 ( 2 ) 197 - 197  2011.02

    CiNii

  • D-12-42 Study of Estimating Face Direction and Forefinger Motion for Manipulating a Virtual Object in a Large Screen : Combining Detection of Face Direction and Particle Filter Based Hand and Finger Tracking

    Wu Ming-Han, Ohya Jun

    Proceedings of the IEICE General Conference   2011 ( 2 ) 145 - 145  2011.02

    CiNii

  • D-11-48 Study of Stereoscopic Enhancement for 3D Real Images by Considering Color Saturation

    Ku Cheng Yu, Ohya Jun, Abe Nobuaki, Kawai Takashi

    Proceedings of the IEICE General Conference   2011 ( 2 ) 48 - 48  2011.02

    CiNii

  • Study of Parameters for Tracking White Road Line by Particle Filter : Study of Recognizing White Road Line from the video Sequence Acquired by the Camera Attached to the Human body for Walking Assistance for Visually Disabled

    TAKAHASHI Shouhei, OHYA Jun

    IEICE technical report   110 ( 418 ) 69 - 74  2011.02

     View Summary

    We propose an algorithm for recognizing white road lines from the video sequence acquired by the camera attached to the body of a walking human, towards the achievement of a walking assistance system for visually disabled. The algorithm consists of a Particle Filter, Principal based white line tracking and a Component Analysis and CLAFIC method based classification for detecting the white line. The parameter values for the particle filter for achieving good tracking performance are experimentally determined. Promising experimental results for tracking the white line are obtained.

    CiNii

  • A Study of Human Motion Recognition Based on Tensor Decomposition using a Multiple Viewpoint Image Sequence

    HORI Takayuki, OHYA Jun, KURUMISAWA Jun

    IEICE technical report   110 ( 414 ) 141 - 146  2011.02

     View Summary

    We propose a Tensor Decomposition based algorithm that recognizes the observed action performed by an unknown person and unknown viewpoint not included in the database. Our previous research aimed motion recognition from one single viewpoint. In this paper, we extend our approach for human motion recognition from an arbitrary viewpoint. To achieve this issue, we set tensor database which are multi-dimensional vectors with dimensions corresponding to human models, viewpoint angles, and action classes. The value of a tensor for a given combination of human silhouette model, viewpoint angle, and action class is the series of mesh feature vectors calculated each frame sequence. To recognize human motion, the actions of one of the persons in the tensor are replaced by the synthesized actions. Then, the core tensor for the replaced tensor is computed. This process is repeated for each combination of action, person, and viewpoint. For each iteration, the difference between the replaced and original core tensors is computed. The assumption that gives the minimal difference is the action recognition result. The recognition results show the validity of our proposed method, the method is experimentally compared with Nearest Neighbor rule. Our proposed method is very stable as each action was recognized with over 75% accuracy.

    CiNii

  • 顔の輪郭映像を用いた会話システムにおける社会的スキルと注視行動に関する考察

    花田研, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    情報処理学会全国大会講演論文集   73rd ( 4 )  2011

    J-GLOBAL

  • サイバーシアターにおける観客が実人間あるいはアバターである場合の観客効果の相違に関する検討

    ZHANG Yong, 大谷淳, 米村俊一, 徳永幸生

    電子情報通信学会大会講演論文集   2011  2011

    J-GLOBAL

  • 人物同士のコミュニケーションを触発する“○”メディアにおける自己開示レベルの検討

    CHEN Li Jen, 大谷淳, 米村俊一, 徳永幸生

    電子情報通信学会大会講演論文集   2011  2011

    J-GLOBAL

  • サイバーシアターにおける観客効果に関する検討~観客が実人間・アバターの場合の特性の違いに関する検討~

    YONG Zhang, 大谷淳, 米村俊一, 米村俊一, 徳永幸生

    電子情報通信学会技術研究報告   110 ( 457(MVE2010 133-174) )  2011

    J-GLOBAL

  • 灰色仮説に基づく照明光色推定法の提案

    川村春美, 米村俊一, 大谷淳, 松浦宣彦

    画像電子学会年次大会予稿集(CD-ROM)   39th  2011

    J-GLOBAL

  • Study on a Method of Estimating Learners’ Subjective Impressions of the Difficulty By Their Learning Actions for e-Learning on Demand

    大川内隆朗, 大谷淳, 米村俊一, 米村俊一, 徳永幸生

    電子情報通信学会技術研究報告   111 ( 332(ET2011 76-87) )  2011

    J-GLOBAL

  • 球面領域判定を用いた一枚の画像からの光源位置推定に関する検討

    立松直倫, XU Rong, 大谷淳, 川村春美, 米村俊一

    情報科学技術フォーラム講演論文集   10th  2011

    J-GLOBAL

  • Study of a Temporal Modified-RANSAC Based Method for the Extraction and 3D Shape Reconstruction of Moving Objects from Dynamic Stereo Images and for Estimating the Camera Pose

    TATEMATSU Naotomo, OHYA Jun

    IEICE technical report   110 ( 324 ) 79 - 84  2010.11

     View Summary

    This paper proposes a Temporal modified RANSAC based method that can discriminate each moving object from the still background, can compute the stereo cameras' egomotion, and can reconstruct 3D structure of each moving object. Our Temporal modified RANSAC segments the detected 3D optical flows into moving objects and background and computes the rotation and translation matrix for each object. Finally, the 3D points are registered using each object's matrix to the initial positions so that the 3D structures of the moving objects and still background are reconstructed. Experiments using multiple moving objects and real stereo sequences demonstrate the effectiveness of our proposed method.

    CiNii

  • A Study on Illuminant Color Estimation based on Gray World Assumption using hue categorization

    KAWAMURA Harumi, YONEMURA Shunichi, OHYA Jun, MATSUURA Norihiko

      2010 ( 6 ) 1 - 6  2010.08

    CiNii

  • G-017 An Improved KFCM Algorithm in the Segmentation of Brain MRIs

    Xu Rong, Ohya Jun

      9 ( 2 ) 563 - 564  2010.08

    CiNii

  • F-033 Interactive Music : Teaching People Guitar Through Gaming Software

    Gaizo John Del, Ohya Jun

      9 ( 2 ) 433 - 434  2010.08

    CiNii

  • J-018 Study of Human Gesture Recognition by Integrating Face and Hand Motion Features

    Luo Dan, Ekenel Hazim Kemal, Ohya Jun

      9 ( 3 ) 505 - 506  2010.08

    CiNii

  • I-037 A Basic Study of Stereophonic Enhancement using Texture Mapping Control

    Ku Cheng Yu, Ohya Jun, Kawai Takashi

      9 ( 3 ) 327 - 328  2010.08

    CiNii

  • H-020 Study of Estimating the Egomotion of the Moving Stereo Cameras in the Environment Including Moving Objects and Reconstructing the Observed Space in 3D

    Tatematsu Naotomo, Xie Yingdi, Ohya Jun

      9 ( 3 ) 177 - 178  2010.08

    CiNii

  • H-030 A Study on Human Motion Recognition Based on Tensor Decomposition using Multiple Video Cameras

    HORI Takayuki, OHYA Jun, KURUMISAWA Jun

      9 ( 3 ) 197 - 198  2010.08

    CiNii

  • Texture Analysis for Food Recognition

    DO Khanh N., OHYA Jun, PISHVA Davar

    IEICE technical report   110 ( 98 ) 69 - 74  2010.06

     View Summary

    This paper studies the effectiveness of texture analysis methods for classifying different food items having the same color. This paper studies two texture analysis methods: gray-level co-occurrence matrix (GLCM) based features and Fourier Transform (FT) based features. We carried out experiments on testing the effectiveness of the two texture features using six different food items, where two food items having three colors: white, red and yellow are used. From the experimental results, it turns out that GLCM features and FT's spatial frequency based features are promising.

    CiNii

  • Texture Analysis for Food Recognition

    DO Khanh N., OHYA Jun, PISHVA Davar

    IEICE technical report   110 ( 97 ) 69 - 74  2010.06

     View Summary

    This paper studies the effectiveness of texture analysis methods for classifying different food items having the same color. This paper studies two texture analysis methods: gray-level co-occurrence matrix (GLCM) based features and Fourier Transform (FT) based features. We carried out experiments on testing the effectiveness of the two texture features using six different food items, where two food items having three colors: white, red and yellow are used. From the experimental results, it turns out that GLCM features and FT's spatial frequency based features are promising.

    CiNii

  • Study of a Car Type Classifying Method using an Automated Frame Segmentation Algorithm : Study of a specific Representation for Convertibles Using Local Maxima Curvature Function Features

    BRIOIS Sebastien, OHYA Jun

    IEICE technical report   109 ( 469 ) 13 - 18  2010.03

     View Summary

    This research paper develops a new method for car type classification using two different car views. We first built a database for each car type based on three geometric ratios. We could obtain a statistical distribution of these parameters values. Dimensions of the unknown car are retrieved using a shape extraction method. In a final part, an index of similarity was defined in order to compare unknown cars geometric proportions with those from the generic models.

    CiNii

  • Study of Tracking Body Parts of a Human Wearing a Multiple-Colored Suit in a Video Sequence for Human Posture Estimation : Study of Occlusion Judgement by Particle Filter

    SHIMIZU Yuta, OHYA Jun, OKAZAKI Jun

    IEICE technical report   109 ( 469 ) 7 - 12  2010.03

     View Summary

    The authors are studying a method for estimating the posture of a person wearing a multilple-colored suit from the video sequence that observes that person. This paper reports our fundamental study of tracking human body parts. Each frame of the video sequence is segmented by a mean-shift algorithm, Using color distribution information of each body part, the body part is tracked by particle filter. Particles' "overflow phenomena", which occurs when the body part starts being occluded, are utilized to judge whether occulusion occurs using an overflow ratio.

    CiNii

  • Study of Estimating a Pan-Tilt Camera's Rotational Parameters from the Video Sequence Acquired by the Camera : Study of Robust Estimation from the Scenes Including Moving Objects

    TOCHIORI Yasufumi, OHYA Jun, EBATA Masaru

    IEICE technical report   109 ( 469 ) 1 - 6  2010.03

     View Summary

    This paper proposes a method for estimating pan-tilt camera's rotational parameters from the video sequence acquired by the camera. In particular, this paper studies a method that deals with moving objects in the observed scene. First, detected and tracked two feature points are projected to the surface of the virtual shphere, whose center is the camera's opytical center and whose radius is the camera's focal length, so that the pan and tilt angles are estimated by using the congruence of the triangles formed by the projected two feature points and optical axis. To estimate the pan-tilt angles accurately, the feature points should be detected from the static objects. This paper discriminates the static objects from the moving objects by judging whether the distance between 2 feature points stays constant or not. The validity of the proposed method is demonstrated through simulation and experiments using real scenes.

    CiNii

  • Study of Extracting Multiple Independently Moving Objects and Estimating the Egomotion from Active Stereo Sequences

    XIE Yingdi, OHYA Jun

    IEICE technical report   109 ( 471 ) 413 - 418  2010.03

     View Summary

    In this paper, we propose a modified RANSAC based method that can detect multiple independently moving objects from the video sequences acquired by active stereo cameras, and can estimate the egomotion of the stereo cameras. From the optical flows that computed by matching the feature descriptors within temporal frames, a modified RANSAC is adopted to extract the valid egomotion models, under which enough homogenous optical flows can be found. Each of the egomotion models corresponds to either the background or one of the independent moving individuals. The optical flows of the background are then utilized to calculate the real egomotion. Experiments using real stereo sequences demonstrate the effectiveness of the proposed method.

    CiNii

  • Study of Recognizing Multiple Persons' Complicated Hand Gestures from an Active Image Sequence : Increasing the Number of Persons and Hand Gesture Categories

    DAN Luo, OHYA Jun

    IEICE technical report   109 ( 471 ) 407 - 412  2010.03

     View Summary

    Recognizing hand gestures from the video sequence acquired by a dynamic camera could be a useful interface for human robot communication. We propose a framework to extract and recognize hand gestures from moving camera images. We improved Human-Following Local Coordinate (HFLC) system to extract hand motion trajectories, and Condensation algorithm and PCA-based algorithm was performed to recognize extracted hand trajectories. Our experimental show the comparison between PCA based approach and Condensation algorithm on 51 kinds of sign language based Japanese and American Sign Language gestures obtained from 7 people.

    CiNii

  • Study of Recognizing Multiple Persons' Complicated Hand Gestures from an Active Image Sequence : Increasing the Number of Persons and Hand Gesture Categories

    DAN Luo, OHYA Jun

    IEICE technical report   109 ( 470 ) 407 - 412  2010.03

     View Summary

    Recognizing hand gestures from the video sequence acquired by a dynamic camera could be a useful interface for human robot communication. We propose a framework to extract and recognize hand gestures from moving camera images. We improved Human-Following Local Coordinate (HFLC) system to extract hand motion trajectories, and Condensation algorithm and PCA-based algorithm was performed to recognize extracted hand trajectories. Our experimental show the comparison between PCA based approach and Condensation algorithm on 51 kinds of sign language based Japanese and American Sign Language gestures obtained from 7 people.

    CiNii

  • Study of Extracting Multiple Independently Moving Objects and Estimating the Egomotion from Active Stereo Sequences

    XIE Yingdi, OHYA Jun

    IEICE technical report   109 ( 470 ) 413 - 418  2010.03

     View Summary

    In this paper, we propose a modified RANSAC based method that can detect multiple independently moving objects from the video sequences acquired by active stereo cameras, and can estimate the egomotion of the stereo cameras. From the optical flows that computed by matching the feature descriptors within temporal frames, a modified RANSAC is adopted to extract the valid egomotion models, under which enough homogenous optical flows can be found. Each of the egomotion models corresponds to either the background or one of the independent moving individuals. The optical flows of the background are then utilized to calculate the real egomotion. Experiments using real stereo sequences demonstrate the effectiveness of the proposed method.

    CiNii

  • D-11-52 Perlin Noise generated Terrain Raycasting on the GPU

    Forstmann Sven, Ohya Jun

    Proceedings of the IEICE General Conference   2010 ( 2 ) 52 - 52  2010.03

    CiNii

  • D-12-17 Hand Gestures' Motion models detection and extraction from moving camera images

    Luo Dan, Ohya Jun

    Proceedings of the IEICE General Conference   2010 ( 2 ) 128 - 128  2010.03

    CiNii

  • D-12-16 Study of Human Action Recognition Using HOG Feature by a Tensor Decomposition Based Approach

    Hori Takayuki, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2010 ( 2 ) 127 - 127  2010.03

    CiNii

  • A-16-9 C.O.T.O. : A Multimedia Choreography System Inspired by Butoh-fu

    Pierce Jennifer, Ohya Jun, Gaizo John Del, Wu Ming-Han

    Proceedings of the IEICE General Conference   2010   227 - 227  2010.03

    CiNii

  • A system that Allows Human Hand's Short Range Movements to Point a Virtual Object in a Large Screen By Tracking the Hand by Computer Vision Based Approach

    WU Ming-Han, Ohya Jun

    FIT 2010, 08     239 - 240  2010

    CiNii

  • D-12-15 Study of Estimating Human Postures Wearing a Multiple-Colored Suit from a Video Sequence : Proposing a Method for Tracking Human Body Parts Using the Particle Filter's "Overflow"

    Shimizu Yuta, Ohya Jun, Okazaki Jun

    Proceedings of the IEICE General Conference     126 - 126  2010

    CiNii

  • D-12-87 Study of Estimating Pan-Tilt Camera's Rotational Parameters from the video Sequence Including Moving Objects

    Tochiori Yasufumi, Ebata Masaru, Ohya Jun

    Proceedings of the IEICE General Conference     198 - 198  2010

    CiNii

  • 輪郭映像を用いた会話システムにおける顔注視行動の分析

    花田研, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    情報処理学会全国大会講演論文集   72nd ( 4 )  2010

    J-GLOBAL

  • “・”プロンプタを用いるコミュニケーションシステムにおけるメッセージ表現の分析

    飯田真也, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    情報処理学会全国大会講演論文集   72nd ( 4 )  2010

    J-GLOBAL

  • サイバーシアターにおいて多数の観客アバターが存在する場合の演技者と観客との間の観客効果に関する基礎的検討

    ZHANG Yong, 大谷淳, 米村俊一, 米村俊一

    電子情報通信学会技術研究報告   110 ( 238(MVE2010 52-73) )  2010

    J-GLOBAL

  • 1チャネルの線画映像コミュニケーションにおける対話ストラテジーと対話特性

    米村俊一, 花田研, 徳永幸生, 大谷淳

    ヒューマンインタフェース学会研究報告集(CD-ROM)   12  2010

    J-GLOBAL

  • 創造的思考を用いたコミュニケーション方法-”・”プロンプタ・システムでの実用について-

    CHEN Li Jen, 米村俊一, 大谷淳, 徳永幸生

    電子情報通信学会技術研究報告   109 ( 471(HIP2009 118-210) )  2010

    J-GLOBAL

  • H-003 Comparison of Several Shape Descriptors in Order to Perform Car Recognition from a Moving Camera

    Briois Sebastien, Ohya Jun, Xie Yingdi

      8 ( 3 ) 101 - 102  2009.08

    CiNii

  • H-029 Fundamental Study of Tracking Human Body Parts for Estimating Postures of a Human Wearing a Multiple Colored Suit

    Shimizu Yuta, Ohya Jun

      8 ( 3 ) 159 - 160  2009.08

    CiNii

  • I-013 Automatic Shape Classification through Descriptor Matching

    Xie Yingdi, Ohya Jun

      8 ( 3 ) 267 - 268  2009.08

    CiNii

  • H-024 Study on performance evaluation of a modified RANSAC based ellipse detection method

    Zhang Yong, Xie Yingdi, Ohya Jun

      8 ( 3 ) 149 - 150  2009.08

    CiNii

  • H-034 Study of Recognizing Hand Gesture from the Video Sequence Acquired by a Moving Camera : Extension to Multiple People's Gestures from a Single Person's Gestures

    Luo Dan, Ohya Jun

      8 ( 3 ) 171 - 172  2009.08

    CiNii

  • I-056 Study of Human Tracking Based on Motion Signature in Multiple Cameras

    Hori Takayuki, Ohya Jun, Kurumisawa Jun

      8 ( 3 ) 371 - 372  2009.08

    CiNii

  • 1A2-C03 A Study Toward Automatically Harvesting Saffron Stigmas Using Image Processing

    TAKANO Akira, OGATA Hiroyuki, MURAMATSU Daigo, OHYA Jun

      2009   "1A2 - C03(1)"-"1A2-C03(4)"  2009.05

     View Summary

    All stigmas of saffrons which are used as a spice or a dye are harvested by hand. However, it is heavy work to gather an enough amount. The purpose of this paper is to examine whether harvesting saffron's stigmas automatically is possible by using image processing. As a process of harvesting saffron stigmas automatically, we conceieved an idea of cutting the root of flower and harvesting stigma which is hanging by turning it down. Features to harvest stigmas are detected by using color information. The feasibility of this method was shown by the experiment.

    CiNii

  • Study on the effect of face parts with line drawing in remote conversation system

    HANADA Ken, YONEMURA Shunichi, TOKUNAGA Yukio, SUGIYAMA Kiyoshi, OHYA Jun

    Human Interface   11 ( 2 ) 131 - 136  2009.05

    CiNii

  • Study of Detecting the Gaze of a Table-game Player from the Facial Images Acquired by Two Directional Cameras

    TANG Huabin, OHYA Jun

    IEICE technical report   108 ( 487 ) 31 - 36  2009.03

     View Summary

    As one of the component technologies that allow computers to compete with humans in table game sessions in real world, the gaze detection by a computer vision approach is important for estimating the other player's strategies. This paper proposes a new gaze detection method. From the frontal face image, the position of the player's nose is detected, and from the image acquired by the lower camera the pupil is detected. A calibration measures the displacement between the pupil's located position and the actual gaze. When the gaze is to be detected, the gaze is estimated by integrating the detection results and calibration. Experiments that simulate Mahjong sessions show the validity of the proposed method.

    CiNii

  • Study of Representing Audiences in a Cyber-Theater

    LIU Qiang, OHYA Jun

    IEICE technical report   108 ( 487 ) 37 - 41  2009.03

    CiNii

  • Creative Thinking Support Communication System : with Random Prompter "●" that Allows Hand-drawing

    CHEN Li Jen, OHYA Jun, YONEMURA Shunichi, TOKUNAGA Yukio

    IEICE technical report   108 ( 487 ) 43 - 48  2009.03

     View Summary

    In our previous research, we have developed a visual tele-communication method using visual Prompter (black circle, "●") image patterns (VP-patterns), and developed our first Communication System with Prompter (CSP1), which allows participants conduct conversation with VP-patterns as visual stimuli for inspiring ideas and supporting creative thinking processes. Although the prototype CSP1 has shown a considerable tendency toward the inspiration of creative ideas by the participants, CSP1 was limited only allows text input. After carefully analyzed and researched, we assumed that creative thinking process might be more enhanced with arbitrary hand-drawing input, and have conducted a paper based simulation, in which the participants can hand-draw arbitrary figures in a paper in which VP-patterns are printed. The simulation results have shown positive outcome, therefore we developed our second networked system, CSP2, which allows hand-drawing.

    CiNii

  • Stakeholder System and Analysis of Extended Producer Responsibility Based on Game Theory

    ZHAO Yiping, OHYA Jun

    IPSJ SIG Notes   2009 ( 32 ) 105 - 112  2009.03

     View Summary

    Extended producer responsibility(EPR) is an effective environmental policy principle in terms of motivating responsible corporate to enhance their product's life-cycle environmental performance. However, on the other hand EPR also challenges greatly the core companies into a complicated and dynamics product life-cycle stakeholder system. In this paper, the structures and features of product life-cycle stakeholder system under EPR is analyzed according to stakeholder theory. Primary and secondary stakeholders are identified followed by their inter-relationships discussed. A two-entity static game model is set up accordingly. Mixed-strategy equilibrium is analyzed and key influential factors are studies in detail. Adaptive decision-making suggestions are made in brief for companies facing EPR in several specific scenarios with the dynamic of the above key factors and information.

    CiNii

  • Ellipse Recognition and its Application toward Medical Imaging via Edge Orientation Feature with A Modified RANSAC

    XIE Yingdi, OHYA Jun

    IEICE technical report   108 ( 485 ) 41 - 46  2009.03

     View Summary

    Shape is one of the most important primitive features of natural objects. As long as shape recognition technology's development, it is applied in wider and wider fields, among which medical imaging is a special area with significant importance. In this paper, we propose a novel processing framework that combines a modified RANSAC with segmenting and reconnectting the spatial connected edge curvelets, in order to achieve robustness s and efficiency. We apply this technology to medical imaging area, more specifically the analysis of erythrocyte, in order to provide assistance with cell recognition and classification.

    CiNii

  • A-15-18 Study of Detecting the Gaze of a Table-game Player from the Facial Images

    Tang Huabin, Ohya Jun

    Proceedings of the IEICE General Conference   2009   254 - 254  2009.03

    CiNii

  • A-15-3 Fundamental Study of Audience Effect in a Cyber-Theater

    Liu Qiang, Ohya Jun

    Proceedings of the IEICE General Conference   2009   239 - 239  2009.03

    CiNii

  • A-14-5 Creative Thinking Support Communication System with Random Prompter "・" that Allows Hand-drawing

    Chen Li Jen, Yonemura Shunichi, Ohya Jun, Tokunaga Yukio

    Proceedings of the IEICE General Conference   2009   232 - 232  2009.03

    CiNii

  • D-12-39 A Local Method for Ellipse Detection

    Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2009 ( 2 ) 148 - 148  2009.03

    CiNii

  • D-12-22 Comparative Study of Image Features Used for Person Identification by a Tensor Decomposition Based Image Processing

    Hori Takayuki, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2009 ( 2 ) 131 - 131  2009.03

    CiNii

  • H-059 Study of Estimating the Rotational Parameters of the Rotatable Camera from the Video Sequence

    Tochiori Yasufumi, Ebata Masaru, Ohya Jun

        235 - 236  2009

    CiNii

  • コミュニケーションにおける会話特性に着目した顔構成要素の役割

    花田研, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    情報処理学会全国大会講演論文集   71st ( 4 )  2009

    J-GLOBAL

  • 自動会話システムにおけるフォントによる感情表現の効果

    野原雄紀, 徳永幸生, 杉山精, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   71st ( 4 )  2009

    J-GLOBAL

  • Study on the effect of face parts with line drawing in remote conversation system

    花田研, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    電子情報通信学会技術研究報告   109 ( 29(WIT2009 1-47) )  2009

    J-GLOBAL

  • “黒丸”プロンプタを用いるコミュニケーションシステムにおける表現特性の分析

    飯田真也, 徳永幸生, 杉山精, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   71st ( 4 )  2009

    J-GLOBAL

  • Study of Gait Identification Based on a Tensor Decomposition Approach

    HORI Takayuki, OHYA Jun, KURUMISAWA Jun

    IEICE technical report   108 ( 363 ) 87 - 92  2008.12

     View Summary

    This paper proposes a Tensor Decomposition Based method for personal identification from silhouette image on time sequential data. Personal Identification tests from gait motion data and the experiments compare with other method. Experiments show the effectiveness of the identification method,

    CiNii

  • Parallel Forward Projection of Large Voxel-Volumes on the GPU

    FORSTMANN Sven, OHYA Jun

    IPSJ SIG Notes   2008 ( 124 ) 11 - 16  2008.12

     View Summary

    We present an efficient SIMD optimized implementation of the parallel voxel-based forward projection algorithm. The algorithm rasterizes RLE encoded volume data in a front to back manner by utilizing a specialized view transform. In addition to the original method, our implementation achieves a significant speedup by utilizing a multi-segment culling mechanism in combination with a binary visibility map. We show that it is possible to uncompress and visualize large volume data directly from GPU memory without streaming data from slower CPU memory..

    CiNii

  • Study of Estimating Postures Including Occluded Postures of a Human Wearing a Multiple Colorsd Suit from Multiple Camera Images

    OKAZAKI Jun, OHYA Jun

    IPSJ SIG Notes. CVIM   2008 ( 82 ) 195 - 200  2008.08

     View Summary

    This paper proposes a method for estimating complicated postures of a human body using color information processing. Difficulties in detecting feature points by analyzing the contour of a human body silhouette lie in the case in which the silhouettes of multiple body parts overlap. Thus, we proposed a method in which a human wears a multiple-colored suit, in which each body part is colored differently, so that the above-mentioned problem can be solved by using color information processing. Conventionally we studied only synthesized human images, not real images. This paper studies a method for extracting each body part by using image segmentation based on mean-shift algorithm. After detecting feature points from the body part extraction result, the proposed method computes the 3D coordinates of each feature point using the principle of triangulation. Experimental results using real human images show the validity of the proposed method.

    CiNii

  • Study of Estimating Postures Including Occluded Postures of a Human Wearing a Multiple Colored Suit from Multiple Camera Images

    OKAZAKI Jun, OHYA Jun

    IEICE technical report   108 ( 199 ) 193 - 198  2008.08

     View Summary

    This paper proposes a method for estimating complicated postures of a human body using color information processing. Difficulties in detecting feature points by analyzing the contour of a human body silhouette lie in the case in which the silhouettes of multiple body parts overlap. Thus, we proposed a method in which a human wears a multiple-colored suit, in which each body part is colored differently, so that the above-mentioned problem can be solved by using color information processing. Conventionally we studied only synthesized human images, not real images. This paper studies a method for extracting each body part by using image segmentation based on mean-shift algorithm. After detecting feature points from the body part extraction result, the proposed method computes the 3D coordinates of each feature point using the principle of triangulation. Experimental results using real human images show the validity of the proposed method.

    CiNii

  • I-034 Study of Detecting Primitive Shapes by Local Optimization Method

    Xie Yingdi, Ohya Jun

      7 ( 3 ) 267 - 268  2008.08

    CiNii

  • H-043 Study of Recognizing Human Actions on Silhouette Feature from a Video Sequence by a Tensor Decomposition Based Approach

    HORI Takayuki, OHYA Jun, KURUMISAWA Jun

      7 ( 3 ) 155 - 156  2008.08

    CiNii

  • H-062 Study of Estimating Postures of a Human Wearing a Multiple Colored Suit from Multiple Camera Images

    Okazaki Jun, Ohya Jun

      7 ( 3 ) 197 - 198  2008.08

    CiNii

  • J-021 The Prototype system of Prompter communication with handwriting and drawing functions

    Chen Li Jen, Forstmann Sven, Ohya Jun, Yonemura Shunichi, Tokunaga Yukio

      7 ( 3 ) 449 - 450  2008.08

    CiNii

  • J-026 Comparative study of Locating the Nose from the Facial Image for Estimating the Face Orientation

    Tang Huabin, Ohya Jun

      7 ( 3 ) 459 - 460  2008.08

    CiNii

  • Study of Recognizing Human Actions Similar to Each Other from a Video Sequence by a Tensor Decomposition Based Approach

    HORI Takayuki, OOYA Jun, KURUMISAWA Jun

    IEICE technical report   108 ( 127 ) 43 - 48  2008.07

     View Summary

    This paper proposes a Tensor Decomposition Based method that can recognize an unknown person's action from a video sequence, where the unknown person is not included in the database (tensor) used for the recognition. The tensor consists of persons, actions and time-series image features. For the observed unknown person's action, one of the actions stored in the tensor is assumed. Using the motion signature obtained from the assumption, the unknown person's actions are synthesized. The actions of one of the persons in the tensor are replaced by the synthesized actions. Then, the core tensor for the replaced tensor is computed. This process is repeated for the actions and persons. For each iteration, the difference between the replaced and original core tensors is computed. The assumption that gives the minimal difference is the action recognition result. Three image features are tested for recognition experiments. Experiments using 15 persons' seven kinds of action show the effectiveness of the recognition method and clarify the useful image features.

    CiNii

  • Study of Stereoscopic Contrast Enhancement Based on Local Contrast and Distance Information

    DING Xiao, OHYA Jun, KAWAI Takashi, ABE Nobuaki

    IEICE technical report   108 ( 127 ) 49 - 53  2008.07

     View Summary

    We proposes a stereoscopic contrast enhancement method that utilizes local gray-level contrast and distance information. A histogram is constructed using the weights of the pixels, which are computed using the local contrast and the distance value at the corresponding pixel in the range image. After equalizing the weighted histogram, another conversion that reflects the weight is performed. Experiments using a stereoscopic display show that the proposed method works better than the conventional histogram equalization and restores degraded images well.

    CiNii

  • A Method for Creative Communication Using Black-Circled Prompters "●" : Analysis of a Method that Allows Users' Freehand Writing

    CHEN Li Jen, OHYA Jun, YONEMURA Shunichi, TOKUNAGA Yukio

    IEICE technical report   107 ( 553 ) 61 - 66  2008.03

     View Summary

    Research into creative thinking-support tools and communication is commonly focused on how to develop and share ideas between participants or with others. In this paper, in order to evaluate the potential variations developmental directions for the visual "●" Prompter Message System (PM-System), we explore a Prompter Message Simulation (PM-Simulation), that uses a printed paper version of visual "●" prompter image patterns (VP-patterns), and allows not only writing texts but also freehand drawing and writing. Experiments have been conducted and analyzed into methods of stimulating the thought process, inspiring ideas, creative and imaginative thinking and enrich self-expression. From the results, a tendency towards inspiring creative ideas by participants has been observed.

    CiNii

  • Effect of character's variable display of Human-Computer Conversation System

    NOHARA Yuuki, TOKUNAGA Yukio, SUGIYAMA Kiyosi, YONEMURA Shunichi, OHYA Jun

      70   289 - 290  2008.03

    CiNii

  • Image Contrast Enhancement Based on Distance Information

    DING Xiao, OHYA Jun, KAWAI Takashi, ABE Nobuaki

    IPSJ SIG Notes. CVIM   2008 ( 27 ) 267 - 270  2008.03

     View Summary

    Image enhancement, especially color image enhancement is a critical area in image processing research. Image contrast enhancement is a Image contrast enhancement is to enhance the contrast of color image so as to make the image more vivid and beautiful. Distance sense is a critical element to present the reality of color images. In order to improve the distance sense, we optimize the histogram equalization(HE) with distance information in enhancement processing and propose a Weighted Histogram Equalization(WHE) method. Near areas of the images are enhanced strongly and far areas weakly. Experimental results show that the proposed method is effective.

    CiNii

  • Visualizing run-length-encoded volume data on modern GPUs

    FORSTMANN Sven, OHYA Jun

    IPSJ SIG Notes. CVIM   2008 ( 27 ) 355 - 358  2008.03

     View Summary

    We present an algorithm for the visualization of large opaque volume data. So far, volume data has mainly been visualized on the GPU using conventional ray-casting or texture-based techniques. However, since the appearance of CUDA, even more complex visualization techniques become possible. We show that large run-length-encoded volume data can efficiently be visualized using CUDA. In our experiments using a NVIDIA 8600 GTS graphics board, even as much as 40.000 x 1024 x 40.000 voxel were visualized in real-time at interactive 40 fps.

    CiNii

  • Comparative Study of Methods for Recognizing Human Actions from a Real Video Sequence

    WANG Weiqing, OHYA Jun

    IPSJ SIG Notes. CVIM   2008 ( 27 ) 103 - 108  2008.03

     View Summary

    This paper explores the effectiveness of using three image features instead of synthesized human motion data by using the real video sequence. We have compared three algorithms that recognize the observed action generated by an unknown person, who is not included in the database. We tested the 4 methods using 3 single image features with 4 human actors and 5 classes of action. In addition, we proved all the three method are useful for the human action recognition.

    CiNii

  • Study of Hand Gesture Recognition from the Video Sequence Acquired by a Dynamic Camera Using Condensation Algorithm

    TAN Ra, OHYA Jun

    IPSJ SIG Notes. CVIM   2008 ( 27 ) 11 - 16  2008.03

     View Summary

    The recognition of human gestures in image sequences is an important and challenging problem that enables human robot co-existence Environment as a human-computer interaction application. There already are many researches working on hand gesture recognition from image sequences acquired by still camera. But using Dynamic Camera is few. Our focus here will be on recognition of gestures from video sequences acquired by a Dynamic camera. We created a very simple and stable extracting method of hand motion model using Human-Following Local Coordinate system(HFLC), and we using condensation algorithm to recognize the hand motion model. We demonstrated 35 kinds of Japanese and American sign languages actions of gesture recognition.

    CiNii

  • Study of Evaluating the Camera's Moving Direction in 3D Object Shape Reconstruction from a Dynamic Camera's Image Based on Monte Carlo Method and Bayes Estimation

    NARISAWA Yu, OHYA Jun, OGATA Hiroyuki

    IPSJ SIG Notes. CVIM   2008 ( 27 ) 181 - 186  2008.03

     View Summary

    In order to reconstruct the entire 3-Dstructure of an object from the video sequence acquired by the camera attached to a mobile robot, one view direction is not sufficient; it is necessary to select the optimal direction at each spot on the robot's path, reconstruct local structures, and integrate the multiple local structures.this paper studies a Monte Cairo method and Bayes Estimationbased method for the local 3-D reconstruction. In addition, we study the volume of information from a view point(the initial frame of the view sequence) is useful for evaluating the robot's moving direction

    CiNii

  • Multiple Shapes Extraction with Guidance to Random Sampling

    XIE Yingdi, OHYA Jun

    Technical report of IEICE. PRMU   107 ( 539 ) 457 - 462  2008.03

     View Summary

    In this paper, we present a novel method that is capable of detecting multiple shapes, including: straight line, circle, ellipse, triangle, rectangle and other polygons. Multiple shapes' detection is realized via analysis of edge orientation curve, which are obtained by tracing the contours of pre-segmented image region, and recording contour pixels' orientations. A modified histogram is generated from the edge orientation curve, for initial shape guess, by two groups: line/polygon, and ellipse/ellipse arc. Based on the primary shape recognition, target objects are precisely estimated through detecting the corresponding equation's parameters.

    CiNii

  • Image Contrast Enhancement Based on Distance Information

    DING Xiao, OHYA Jun, KAWAI Takashi, ABE Nobuaki

    IEICE technical report   107 ( 539 ) 267 - 270  2008.03

     View Summary

    Image enhancement, especially color image enhancement is a critical area in image processing research. Image contrast enhancement is a Image contrast enhancement is to enhance the contrast of color image so as to make the image more vivid and beautiful. Distance sense is a critical element to present the reality of color images. In order to improve the distance sense, we optimize the histogram equalization (HE) with distance information in enhancement processing and propose a Weighted Histogram Equalization (WHE) method. Near areas of the images are enhanced strongly and far areas weakly. Experimental results show that the proposed method is effective.

    CiNii

  • Study of Evaluating the Camera's Moving Direction in 3D Object Shape Reconstruction from a Dynamic Camera's Image Based on Monte Carlo Method and Bayes Estimation

    NARISAWA Yu, OHYA Jun, OGATA Hiroyuki

    IEICE technical report   107 ( 539 ) 181 - 186  2008.03

     View Summary

    In order to reconstruct the entire 3-Dstructure of an object from the video sequence acquired by the camera attached to a mobile robot, one view direction is not sufficient; it is necessary to select the optimal direction at each spot on the robot's path, reconstruct local structures and integrate the multiple local structures.this paper studies a Monte Calro method and Bayes Estimationbased method for the local 3-D reconstruction. In addition, we study the volume of information from a view point (the initial frame of the view sequence) is useful for evaluating the robot's moving direction

    CiNii

  • Study of Hand Gesture Recognition from the Video Sequence Acquired by a Dynamic Camera Using Condensation Algorithm

    TAN Ra, OHYA Jun

    IEICE technical report   107 ( 539 ) 11 - 16  2008.03

     View Summary

    The recognition of human gestures in image sequences is an important and challenging problem that enables human robot co-existence Environment as a human-computer interaction application. There already are many researches working on hand gesture recognition from image sequences acquired by still camera. But using Dynamic Camera is few. Our focus here will be on recognition of gestures from video sequences acquired by a Dynamic camera. We created a very simple and stable extracting method of hand motion model using Human-Following Local Coordinate system (HFLC), and we using condensation algorithm to recognize the hand motion model. We demonstrated 35 kinds of Japanese and American sign languages actions of gesture recognition.

    CiNii

  • D-12-57 Fundamental Study of Tracking a Ball Moving Fast in Sports Scenes

    Liu Qiang, Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2008 ( 2 ) 188 - 188  2008.03

    CiNii

  • D-12-7 Study of Detecting Hand Gestures' Trajectories by Human-Following Local Coordinate System from the Video Sequence Acquired by a Dynamic Camera

    Ra tan, Ohya Jun

    Proceedings of the IEICE General Conference   2008 ( 2 ) 138 - 138  2008.03

    CiNii

  • D-12-36 Study of Evaluating the Camera's Moving Direction in 3D Object Shape Reconstruction from a Dynamic Camera

    NARISAWA Yu, OHYA Jun, OGATA Hiroyuki

    Proceedings of the IEICE General Conference   2008 ( 2 ) 167 - 167  2008.03

    CiNii

  • D-15-32 Design and Implementation of Video Skimming for E-Learning

    OHKAWAUCHI Takaaki, OHYA Jun

    Proceedings of the IEICE General Conference   2008 ( 1 ) 226 - 226  2008.03

    CiNii

  • D-11-68 Study of Effective Features for Recognizing Human Actions from a Real Video Sequence

    Wang Weiqing, Ohya Jun

    Proceedings of the IEICE General Conference   2008 ( 2 ) 68 - 68  2008.03

    CiNii

  • D-12-115 Detecting Ellipses from Image by Randomized Hough Transform with Edge Orientation Difference Curve Based Clustering

    Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2008 ( 2 ) 246 - 246  2008.03

    CiNii

  • A-14-12 A Caricatured Portrait Generate System Which Can Categorize and Describe Facial Feature : Automatic Caricatured Portrait Generate System :ACGS

    Enomoto Makoto, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2008   249 - 249  2008.03

    CiNii

  • A-14-5 A Fundamental Study of "Bi-Conversation" Using "●" Communication Method

    Chen Li Jen, Yonemura Shunichi, Ohya Jun, Tokunaga Yukio

    Proceedings of the IEICE General Conference   2008   242 - 242  2008.03

    CiNii

  • A-15-22 Fundamental Study of Gaze Detection from the Nose Position Using NIR-Image

    Tang Huabin, Ohya Jun, Ohkawauchi Takaaki

    Proceedings of the IEICE General Conference   273   273 - 273  2008

    CiNii

  • D-12-3 Study of Mean Shift Algorithm Based Segmentation for Real Color Images for a Human Wearing a Multipe Colored Suit

    Okazaki Jun, Ohya Jun, Gyo Huimin

    Proceedings of the IEICE General Conference     134 - 134  2008

    CiNii

  • “○”プロンプタを用いたコミュニケーションシステムの会話特性

    飯田真也, 徳永幸生, 杉山精, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   70th ( 4 )  2008

    J-GLOBAL

  • Study of Different Image Features Used for Recognizing Unknown Person's Activity Based on Tensor Decomposition Approach

    IRAWAN Acep, XIE Yingdi, OHYA Jun

    IPSJ SIG Notes. CVIM   2007 ( 87 ) 239 - 244  2007.09

     View Summary

    This paper proposes a method for recognizing an unknown person' action from a video sequence, where the unknown person is defined as a person who is not included in the database (tensor) used for the recognition process. We have developed a tensor decomposition based method, but that method used only motion capture data. The main motivation of this paper is to replace the motion capture data by image feature data and clarify which image features are useful. This paper tests three image features as well as their combinations. Experimental results using synthesized image sequences present the effectiveness of the method and clarify the useful image features.

    CiNii

  • Comparative Study of Segmentation in Monochrome Facial Images: Snake Pedals and Template Matching

    RAMIREZ Carolina, PEREZ Claudio, OHYA Jun

    IPSJ SIG Notes. CVIM   2007 ( 87 ) 167 - 172  2007.09

     View Summary

    The snake pedal model is a rather new active contour algorithm used for segmentation in 2D and 3D digital images, and has been employed with good results in medical imaging. However, its use on other typical images, like facial images, has not been studied deeply. In this paper we evaluate the snake pedal model applied to face segmentation, and more specifically, to parameter extraction, i.e., face aspect ratio, face area and position. We compare its results with the ones of a template matching method. Results indicate that the snake pedal model outperforms the template matching model in face area estimation.

    CiNii

  • A Modified RANSAC Based Method for Extracting Primitive Shapes from Image

    XIE Yingdi, OHYA Jun

    IPSJ SIG Notes. CVIM   2007 ( 87 ) 71 - 76  2007.09

     View Summary

    Primitive shapes extraction in image processing, such as line and circle, is of fundamentally importance. In this paper, we propose a modified ransac based shape detection method. Through a repetitive process that randomly selecting points and validity evaluation, feature models could be successfully obtained. In the experimental results, the effectiveness of this method is demonstrated.

    CiNii

  • A Modified RANSAC Based Method for Extracting Primitive Shapes from Image

    XIE Yingdi, OHYA Jun

    IEICE technical report   107 ( 206 ) 71 - 76  2007.08

     View Summary

    Primitive shapes extraction in image processing, such as line and circle, is of fundamentally importance. In this paper, we propose a modified ransac based shape detection method. Through a repetitive process that randomly selecting points and validity evaluation, feature models could be successfully obtained. In the experimental results, the effectiveness of this method is demonstrated.

    CiNii

  • Study of Different Image Features Used for Recognizing Unknown Person's Activity Based on Tensor Decomposition Approach

    IRAWAN Acep, XIE Yingdi, OHYA Jun

    IEICE technical report   107 ( 207 ) 239 - 244  2007.08

     View Summary

    This paper proposes a method for recognizing an unknown person' action from a video sequence, where the unknown person is defined as a person who is not included in the database (tensor) used for the recognition process. We have developed a tensor decomposition based method, but that method used only motion capture data. The main motivation of this paper is to replace the motion capture data by image feature data and clarify which image features are useful. This paper tests three image features as well as their combinations. Experimental results using synthesized image sequences present the effectiveness of the method and clarify the useful image features.

    CiNii

  • Study of Different Image Features Used for Recognizing Unknown Person's Activity Based on Tensor Decomposition Approach

    IRAWAN Acep, XIE Yingdi, OHYA Jun

    IEICE technical report   107 ( 206 ) 239 - 244  2007.08

     View Summary

    This paper proposes a method for recognizing an unknown person' action from a video sequence, where the unknown person is defined as a person who is not included in the database (tensor) used for the recognition process. We have developed a tensor decomposition based method, but that method used only motion capture data. The main motivation of this paper is to replace the motion capture data by image feature data and clarify which image features are useful. This paper tests three image features as well as their combinations. Experimental results using synthesized image sequences present the effectiveness of the method and clarify the useful image features.

    CiNii

  • A Modified RANSAC Based Method for Extracting Primitive Shapes from Image

    XIE Yingdi, OHYA Jun

    IEICE technical report   107 ( 207 ) 71 - 76  2007.08

     View Summary

    Primitive shapes extraction in image processing, such as line and circle, is of fundamentally importance. In this paper, we propose a modified ransac based shape detection method. Through a repetitive process that randomly selecting points and validity evaluation, feature models could be successfully obtained. In the experimental results, the effectiveness of this method is demonstrated.

    CiNii

  • J-024 Tensor Decomposition Framework For Recognizing an Unknown Person's Action From A Video Sequence Using Image Features

    Irawan Acep, Xie Yingdi, Ohya Jun

      6 ( 3 ) 443 - 444  2007.08

    CiNii

  • J-041 Topic support in automatic conversation system in background image

    NOHARA Yuki, TOKUNAGA Yukio, SUGIYAMA Kiyosi, YONEMURA Shunichi, OHYA Jun

      6 ( 3 ) 481 - 482  2007.08

    CiNii

  • I-066 Low Level Feature Detection based on Modified Ransac

    Xie Yingdi, Ohya Jun

      6 ( 3 ) 345 - 346  2007.08

    CiNii

  • J-005 The Relationship between Caricatured Portrait and Dual Coding Theory in The Case of Face Recognition : Application to Caricatured Portrait Generate System Using Computer

    Enomoto Makoto, Ohya Jun, Kurumisawa Jun

      6 ( 3 ) 397 - 398  2007.08

    CiNii

  • H-024 Validation of the Snake Pedal Model applied to Face Segmentation in Monochromatic Images

    Ramirez Carolina, Perez Claudio, Ohya Jun

      6 ( 3 ) 57 - 58  2007.08

    CiNii

  • H-008 Study of Constructing a Robot Platform for Tracking Moving Objects from Mobile Video Sequence

    Ra Tan, Xie Yingdi, Ohya Jun

      6 ( 3 ) 21 - 22  2007.08

    CiNii

  • H-058 Study of Foreground-Background Contrast Enhancement in Color Image

    DING Xiao, OHYA Jun

      6 ( 3 ) 139 - 140  2007.08

    CiNii

  • H-056 A Study of Monte Carlo and Bayes Estimation Based Method for Reconstructing 3D Objects in Consideration of Vertical Movements of the Camera

    NARISAWA Yu, OGATA Hiroyuki, OHYA Jun

      6 ( 3 ) 133 - 134  2007.08

    CiNii

  • O-009 Consideration of Image Processing for Recognition Method of Customer Behavior at the Time of Choice in the Store

    Hori Takayuki, Kurumisawa Jun, Ohya Jun

      6 ( 4 ) 453 - 454  2007.08

    CiNii

  • Communication analysis on visual support using line drawing expression

    YONEMURA Shunichi, YOSHIDA Tohru, TOKUNAGA Yukio, OHYA Jun

    Human Interface   9 ( 2 ) 157 - 162  2007.05

    CiNii

  • Communication analysis on visual support using line drawing expression

    YONEMURA Shunichi, YOSHIDA Tohru, TOKUNAGA Yukio, OHYA Jun

    IEICE technical report   107 ( 60 ) 157 - 162  2007.05

     View Summary

    Novices have little technical skill and so are best supported by verbal explanations by experts in conjunction with visual information such as pictures of computer screens. Video telephone systems are becoming more prevalent and are a logical infrastructure on which to build a novice support service. Unfortunately, there is a lot of resistance to such systems since most users feel uncomfortable with showing their faces and dwellings to strangers, the experts. Key problems are the loss of privacy and the feeling of breakdown in security. This paper proposes a system that automatically deforms the images transmitted in both directions os as to achieve two goals: easing the concerns of users and ensuring good enough support. We examine here the effect of visual information quality on the effectiveness and impression of a collaboration between a novice user and an operator.

    CiNii

  • A STUDY OF A COMPUTER MEDIATED COMMUNICATION VIA THE "●"PROMPT SYSTEM : Analysis of the affects on the stimulation of thought processes and the inspiration of creative ideas

    CHEN Li Jen, HARADA Nobuyuki, YONEMURA Shunichi, OHYA Jun, TOKUNAGA Yukio

    IEICE technical report   106 ( 609 ) 27 - 30  2007.03

     View Summary

    Research into thinking-support tools is commonly focused on how to develop and share ideas between participants or with others. In this paper, we propose and develop a communication system that stimulates the thought processes and inspires the creative ideas of participants by using a visual "●" prompt within the framework of a communication pallet. Experiments have been conducted into methods of stimulating the thought process and inspiring ideas during conversation and the results have been analyzed. From the results, a tendency towards inspiring creative ideas by participants has been observed.

    CiNii

  • Study of Image Contrast Enhancement Based on Shadow and Non-shadow Area

    DING Xiao, OHYA Jun

    IEICE technical report   106 ( 608 ) 53 - 58  2007.03

     View Summary

    Image enhancement, especially color image enhancement is a critical area in image processing research. Image contrast enhancement is to enhance the contrast of color image so as to make the image more vivid and veautiful. Shadow is an integral part of color images but its features are different from non-shadow area so that the processing of shadow and non-shadow should also be different. We identify cast shadows by shadow detection method and divide the color image into shadow area and non-shadow area. Then apply color enhancement method in HSV color space in non-shadow areasand shadow areas respectively. Experimental results show that our proposed method based on shadow and non-shadow area can successfully enhance color images.

    CiNii

  • Adaptive Optimization to Modified Ransac Based Motion Segmentation

    XIE Yingdi, OHYA Jun

    IEICE technical report   106 ( 608 ) 83 - 86  2007.03

     View Summary

    In this paper, adaptive optimization methods to video sequence, which will apply to modified ransac based multiple independent motion segmentation on is discussed. Give that our system could run real-time computation, the speed continuity is a strong constraint to the following three points: 1) optical flow sampling 2) adaptive thresholds 3) motion continuity of both the camera and independent moving objects. From the next section, we discuss the corresponding solution to the above expecting improvements.

    CiNii

  • Comparative Study of Marker-based Detection and Tracking of PC in the Video Sequence Acquired by the Mobile Camera for Visual Call Center Applications

    ZHANG Qi, OHYA Jun, YONEMURA Shunichi

    IEICE technical report   106 ( 608 ) 77 - 82  2007.03

     View Summary

    The call centers in which video communication is available, visual call centers, are supposed to be an efficient way of PC troubleshooting. Users take the video of their computer with handy cameras and send the video to the operators. The operators can confirm the matter directly so that they can give more reasonable advice to users. However, viewing the video directly has the threat of privacy violation. To overcome this problem, a method that can detect computer from the video and eliminate other objects is needed. In this paper, a marker-based PC detection method is proposed and several typical tracking methods are compared in order to fine out the appropriate ones. The accuracy and robustness of both the marker-based methods and the tracking methods are shown by experimental results.

    CiNii

  • D-12-149 Marker-based PC Detection Using Sample Matching

    Zhang Qi, Ohya Jun, Yonemura Shunichi

    Proceedings of the IEICE General Conference   2007 ( 2 ) 265 - 265  2007.03

    CiNii

  • D-12-25 Stability Improvement for Motion Segmentation based on Active Stereo Cameras

    Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2007 ( 2 ) 141 - 141  2007.03

    CiNii

  • D-12-123 Consideration of Image Processing Based on Observation Method to Recognition of the Customer Behavior

    Hori Takayuki, Kurumisawa Jun, Ohya Jun

    Proceedings of the IEICE General Conference   2007 ( 2 ) 239 - 239  2007.03

    CiNii

  • D-12-115 Study of Contrast Enhancement Based on Shadow Detection

    Xiao Ding, Ohya Jun

    Proceedings of the IEICE General Conference   2007 ( 2 ) 231 - 231  2007.03

    CiNii

  • D-12-122 Implementation of a Video Sequence Based System for Recognizing an Unknown Person's Gesture by a Tensor Decomposition Based Approach

    Irawan Acep, Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2007 ( 2 ) 238 - 238  2007.03

    CiNii

  • A-14-10 Hierarchical Structure for Memories of Faces in Analytic Representation Area and Computer Simulation : Application to Caricatured Portrait Generate System Using Computer

    Enomoto Makoto, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2007   279 - 279  2007.03

    CiNii

  • A-14-9 A Study of Communication Prompted by "・" Communication System : Approach on how prompter effects and inspires ideas and thoughts

    Chen Li Jen, Harada Nobuyuki, Ohya Jun, Yonemura Shunichi, Tokunaga Yukio

    Proceedings of the IEICE General Conference   2007   278 - 278  2007.03

    CiNii

  • 顔の輪郭画像を用いた会話システムにおける会話特性

    吉田亨, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   69th ( 4 )  2007

    J-GLOBAL

  • “○”プロンプタを用いるコミュニケーションシステム

    原田宣之, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   69th ( 4 )  2007

    J-GLOBAL

  • 線画映像による遠隔サポート-触覚インタラクションの展望-

    米村俊一, 吉田亨, 徳永幸生, 大谷淳

    人間工学   43  2007

    J-GLOBAL

  • Independent motion areas extraction from a modified RANSAC mechanism

    XIE Yingdi, OHYA Jun

    IEICE technical report   106 ( 301 ) 13 - 16  2006.10

     View Summary

    Independent motion area detection is one of the active researched topics among computer vision. The computation could be achieved from seperating background and also different independent motion areas. However, the estimation within a scene with independent motion area is difficult for its uncertainty of optical flow selection. In this paper, we propse a method of different independent motion areas from a modified RANSAC mechanism..

    CiNii

  • Object discrimination using Sequential estimation of surface materials

    IKEZAKI Taro, TANAKA Hiroaki, OGATA Hiroyuki, OHYA Jun

    FAN Symposium : Intelligent System Symposium-fuzzy, AI, neural network applications technologies   16   223 - 224  2006.09

     View Summary

    It is important for robots to recognize unregistered objects when working in general environments like home or office. In such a case, the information of shape and material may play an important role to recognize them. In this research, a sequential method of inferring object's surface material using brightness data from a series of images is examined.

    CiNii

  • Study of Real-time Marker-based PC Detection for Visual Call Center Applications

    ZHANG Qi, OHYA Jun, YONEMURA Shunichi

    IPSJ SIG Notes. CVIM   2006 ( 93 ) 227 - 232  2006.09

     View Summary

    Nowadays, majority of PC producers provide their product support services via visual call centers. At times, users cannot describe their problems correctly which causes waste of time. We propose a new solution for PC based visual call center applications. The video images taken by the user's portable camera are used for smooth communications between the user and the operator at the visual call center. To protect the user's privacy, only the PC in the video images should be visible to the operator. This paper proposes a marker based method for detecting the PC in real-time. The design of the markers that are attached to the PC are discussed and studied. A modified Chamfer Matching Algorithm is used for detecting the markers. Experimental results demonstrate the effectiveness and efficiency of the proposed method.

    CiNii

  • A Modified RANSAC mechanism:Multiple models extraction algorithm

    XIE Yingdi, OHYA Jun

    IPSJ SIG-CVIM : Computer Vision and Image Media   2006 ( 93 ) 159 - 166  2006.09

     View Summary

    As one of the most frequently used regression methods, RANSAC is advanced in its effectiveness and efficiency, but RANSAC cannot extract multiple models due to its exclusivity. To extract multiple models, this paper proposes a new regression method, which is a modified version of RANSAC. To fit a model to data points, a labeling process classifies each data point into proper inlier, quasi-inlier or outlier. The model is obtained from the proper and quasi inliers. After eliminating the proper inliers, another model fitting is performed. These operations are repeated till no more model is fitted. The effectiveness of the proposed method is shown by experiments on extracting multiple lines from images.

    CiNii

  • Study of Color Image Enhancement Based on Wavelet Analysis

    DING Xiao, OHYA Jun

    IPSJ SIG Notes. CVIM   2006 ( 93 ) 35 - 40  2006.09

     View Summary

    Image enhancement, especially color image enhancement is a critical area in image processing research. The wavelet analysis is widely used in signal processing and image processing. By using wavelet transform, wavelet inverse transform and an enhancement processing between the two transforms, we propose a color image enhancement method in HSV color space. More specifically, Wavelet transform is applied to the luminance component. The coefficients for the approximation components are enhanced by a gray-level image contrast enhancement technique, and then using the enhanced coefficients, the inverse Wavelet transform is performed. To the saturation component, the histogram equalization is applied. Promising results are obtained by applying the proposed method to some color images. The experimental result showed that the enhancement method can improve the performance of color images.

    CiNii

  • A Modified RANSAC mechanism : Multiple models extraction algorithm

    XIE Yingdi, OHYA Jun

    IEICE technical report   106 ( 230 ) 25 - 32  2006.09

     View Summary

    As one of the most frequently used regression methods, RANSAC is advanced in its effectiveness and efficiency, but RANSAC cannot extract multiple models due to its exclusivity. To extract multiple models, this paper proposes a new regression method, which is a modified version of RANSAC. To fit a model to data points, a labeling process classifies each data point into proper inlier, quasi-inlier or outlier. The model is obtained from the proper and quasi inliers. After eliminating the proper inliers, another model fitting is performed. These operations are repeated till no more model is fitted. The effectiveness of the proposed method is shown by experiments on extracting multiple lines from images.

    CiNii

  • Study of Real-time Marker-based PC Detection for Visual Call Center Applications

    ZHANG Qi, OHYA Jun, YONEMURA Shunichi

    IEICE technical report   106 ( 230 ) 93 - 98  2006.09

     View Summary

    Nowadays, majority of PC producers provide their product support services via visual call centers. At times, users cannot describe their problems correctly which causes waste of time. We propose a new solution for PC based visual call center applications. The video images taken by the user's portable camera are used for smooth communications between the user and the operator at the visual call center. To protect the user's privacy, only the PC in the video images should be visible to the operator. This paper proposes a marker based method for detecting the PC in real-time. The design of the markers that are attached to the PC are discussed and studied. A modified Chamfer Matching Algorithm is used for detecting the markers. Experimental results demonstrate the effectiveness and efficiency of the proposed method.

    CiNii

  • Study of Color Image Enhancement Based on Wavelet Analysis

    DING Xiao, OHYA Jun

    IEICE technical report   106 ( 229 ) 35 - 40  2006.09

     View Summary

    Image enhancement, especially color image enhancement is a critical area in image processing research. The wavelet analysis is widely used in signal processing and image processing. By using wavelet transform, wavelet mverse transform and an enhancement processing between the two transforms, we propose a color image enhancement method in HSV color space. More specifically, Wavelet transform is applied to the luminance component. The coefficients for the approximation components are enhanced by a gray-level image contrast enhancement technique, and then using the enhanced coefficients, the inverse Wavelet transform is performed. To the saturation component, the histogram equalization is applied. Promising results are obtained by applying the proposed method to some color images. The experimental result showed that the enhancement method can improve the performance of color images.

    CiNii

  • K_089 Analysis of Communication Prompted by Black "●" Communication System : Effects of Emotional Emphasis in Comparison with E-mails and Messengers

    Chen Li Jen, Harada Nobuyuki, Ohya Jun, Tokunaga Yukio, Yonemura Shunichi

      5 ( 3 ) 577 - 578  2006.08

    CiNii

  • K_008 Hierarchical Structure for Memories of Faces in Analytic Representation Area and Computer Simulation : Application to Caricatured Portrait Generate System Using Computer

    Enomoto Makoto, Ohya Jun, Kurumisawa Jun

      5 ( 3 ) 383 - 384  2006.08

    CiNii

  • I_015 Optimization of 3D Egomotion Computation in Image Sequence

    Xie Yindi, Ohya Jun

      5 ( 3 ) 33 - 34  2006.08

    CiNii

  • I_074 Study of Color Contrast Enhancement Based on Wavelet Transform

    Xiao Ding, Ohya Jun

      5 ( 3 ) 179 - 180  2006.08

    CiNii

  • J_019 Study of Reducing Computation Costs and Memory Usage for Articulated Elastic OBject Models

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

      5 ( 3 ) 233 - 234  2006.08

    CiNii

  • I_053 Object Detection by Using Marker Assistance

    Zhang Qi, Ohya Jun, Yonemura Shunichi

      5 ( 3 ) 133 - 134  2006.08

     View Summary

    This paper proposes a solution for PC troubleshooting which can be used in the visual call center. The operator and user communicate not only by audio channel, but also by visual channel. To protect users' privacy, a method for eliminating objects except the users' PC from each frame of the video stream is needed. First, the marker that is to be attached to the PC is designed so that the marker is rotation-invariant and well-contrasted. After detecting candidate regions for the markers in each frame, a modified Chamfer Matching Algorithm is applied so that the markers are detected. Experimental results demonstrate the effectiveness of the proposed method.

    CiNii

  • The Study on Extraction of Emotional Features from Analyzing Human Body Motion

    KOBAYASHI Yuichi, OHYA Jun

    IEICE technical report   106 ( 143 ) 97 - 102  2006.06

     View Summary

    In order to analyze and extract differences between human body actions with different emotion, gait actions which actors or actresses played each emotion were recorded by means of a motion capture system. Each time sequence was analyzed by analysis of velocity and acceleration, and moreover by analyzing state transfer based on a simple description which can describe time variate posture. Experimental results shows that our method can adequately classify each emotional gait motion.

    CiNii

  • ANALYSIS OF HUMAN-TO-HUMAN COMMUNICATIONS PROMPTED BY THE "●" COMMUNICATION SYSTEM : Comparing "with Prompters" and "without Prompters"

    CHEN Li Jen, OHYA Jun, TOKUNAGA Yukio, YONEMURA Shunichi

    IEICE technical report   105 ( 683 ) 31 - 36  2006.03

     View Summary

    This paper proposes a new communication system using a black "●" prompter to inspire and encourage conversation. Experiments related to the usage of the prompter have been conducted and analyzed. Conversation experiments, both with and without a prompter, have also been performed. Following an analysis of the experimental data, it was discovered that the communication system using the prompter was supportive in expressing information other than text. Through this research and study, it can be seen that a communication method using the "●" prompter can potentially be used as an effective conversation tool.

    CiNii

  • Study of A Caricatured Portrait Generate System Which Can Categorize and Describe Facial Feature

    ENOMOTO Makoto, OHYA Jun, KURUMISAWA Jun

    IEICE technical report   105 ( 683 ) 19 - 24  2006.03

     View Summary

    Viewing the process of drawing caricatured portrait by artist myself introspectively, and it to be a computer program called 'Automatic caricatured portrait generate systems'.

    CiNii

  • Study of the Creation Processes-Based System for Exhibiting Artistic Works

    SRICHOMPOO Ton, OHYA Jun, KURUMISAWA Jun, ENOMOTO Makoto

    IEICE technical report   105 ( 683 ) 61 - 66  2006.03

     View Summary

    Art museum exhibits the artworks in term of inspiration and education. It is a relaxation place for one who loves and appreciates art. It is also a place where designers and artists seek for inspiration for their works. However, not many people have visited the art museum, because not many people can understand or appreciate the beauty or meaning of the artworks. One solution for these problems could be to exhibit not only the final work but also its creation process. This paper addresses an issue of how to construct a system that can exhibit artistic works' creation processes.

    CiNii

  • Study of Motion Segmentation from Dynamic Stereo Images Acquired by Active Stereo Cameras

    XIE Yingdi, OHYA Jun

    IEICE technical report   105 ( 689 ) 1 - 6  2006.03

     View Summary

    In this paper, we aim at achieving a real-time method for segmenting moving objects from dynamic stereo images acquired by the stereo cameras attached to a movable platform such as a robot. Although some algorithms are proposed to this problem, the central problem faced by all motion-based techniques is that, in general, it is difficult to estimate 3D motion from flow at one step, and it is difficult in achieving computational efficiency. We address this problem by implementing a computationally efficient regression method: RANSAC to calculate the ego-motion of the camera, and select a robust and fast stereo matching method to acquire the 3D structure of scene.

    CiNii

  • Basic Study for Recognition method of Customer Behavior : Judgment of Flames which Human Existence and Recognition of Interaction Between Human and an Object Using by Color Information

    HORI Takayuki, OHYA Jun, KURUMISAWA Jun

    IEICE technical report. Image engineering   105 ( 689 ) 7 - 8  2006.03

     View Summary

    I was aimed at recognition method of customer behavior at the time of the purchasing and action such as comparison in a store. I do not treat recognition of human behavior directly here, but I detect a time when a person intervenes in as a key frame and recognize a state of the scene by comparing the key frames. If I find a difference between the key frame, I recognize it as a human contact with an object.

    CiNii

  • Study of a System Supporting Fire Extinguishing Activities by Using Sensor Networks : Study of Recognizing the Situation of Fire by Image Processing Technologies

    NOGAI Masahiro, OHYA Jun

    IEICE technical report   105 ( 674 ) 15 - 20  2006.03

     View Summary

    Towards the realization of a system that could support fire extinguishing activities by combining sensor networks and image processing technologies, this paper studies an image processing based method for recognizing the situation of fire. More specifically, how to recognize behaviors of smoke is studied. To detect smoke's behaviors, in general simple background subtractions do not work Therefore, this paper proposes an image segmentation method that recursively partitions the image using a feature based on gray-level variances. Experimental results using real video sequences of fire demonstrate the effectiveness of the proposed algorithm.

    CiNii

  • Study of a System Supporting Fire Extinguishing Activities by Using Sensor Networks- Study of Recognizing the Situation of Fire by Image Processing Technologies -

    NOGAI Masahiro, OHYA Jun

    IPSJ SIG Notes. CVIM   2006 ( 25 ) 187 - 192  2006.03

     View Summary

    Towards the realization of a system that could support fire extinguishing activities by combining sensor networks and image processing technologies, this paper studies an image processing based method for recognizing the situation of fire. More specifically, how to recognize behaviors of smoke is studied. To detect smoke's behaviors, in general simple background subtractions do not work. Therefore, this paper proposes an image segmentation method that recursively partitions the image using a feature based on gray-level variances. Experimental results using real video sequences of fire demonstrate the effectiveness of the proposed algorithm.

    CiNii

  • D-12-57 Study of Basian Sequential 3D Shape Restoration from an Active Camera : Proposing a Method that Does Not Require Feature Point Tracking

    NARISAWA Yu, OGATA Hiroyuki, OHYA Jun

    Proceedings of the IEICE General Conference   2006 ( 2 ) 189 - 189  2006.03

    CiNii

  • D-12-105 Basic Study for Recognition Method of Existence of an object Using by Color Information

    Hori Takayuki, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2006 ( 2 ) 237 - 237  2006.03

    CiNii

  • D-12-100 Study of Motion Segmentation from Dynamic Stereo Images Acquired by Active Stereo Cameras

    Xie Yingdi, Ohya Jun

    Proceedings of the IEICE General Conference   2006 ( 2 ) 232 - 232  2006.03

    CiNii

  • D-12-76 A Study of Computer Vision Based Recognition of Smoke's Behaviors towards the Realization of Fire Extinguishing Systems Using Sensor Networks

    Nogai Masahiro, Ohya Jun

    Proceedings of the IEICE General Conference   2006 ( 2 ) 208 - 208  2006.03

    CiNii

  • A-15-4 Proposal of A Caricatured Portrait Generate System Which Can Categorize and Describe Facial Feature

    Enomoto Makoto, Ohya Jun, Kurumisawa Jun

    Proceedings of the IEICE General Conference   2006   240 - 240  2006.03

    CiNii

  • A-15-12 Study of Designing a Prototype System for Exhibiting Artistic Works with Their Creation Processes

    Srichompoo Ton, Ohya Jun, Kurumisawa Jun, Enomoto Makoto

    Proceedings of the IEICE General Conference   2006   248 - 248  2006.03

    CiNii

  • A-14-2 A Study of Communication Prompted by the "●" Communication System : Comparing "with Prompters" with "without Prompters"

    Chen Li Jen, Ohya Jun, Tokunaga Yukio, Yonemura Shunichi

    Proceedings of the IEICE General Conference   2006   233 - 233  2006.03

    CiNii

  • D-12-126 Construction Method for Character Models behaving like an Elastic Object

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

    Proceedings of the IEICE General Conference   2006 ( 2 ) 258 - 258  2006.03

    CiNii

  • Propose of Construction Method for Elastic Character Models that Enable Interactive Manipulation by Using Two or More Elastic Objects

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

    IPSJ SIG Notes   2006 ( 18 ) 81 - 86  2006.02

     View Summary

    This paper proposes an articulated 3D elastic object model that allows interactive deformations including expansions and contractions. The proposed model consists of multiple elastic object parts, and is surrounded by an exoskeleton. Adjacent elastic object parts are connected by a joining surface, which corresponds to a joint. An elastic object part consists of a bone and elastic object primitives. The exoskeleton is used for interactive deformations; i.e., according to the displacement of the vertex of the exoskeleton, the bone's deformation including expansion and contraction is computed by inverse-kinematics etc. Based on the bone's deformation, the deformations of the elastic object primitives and joining surfaces are computed by the boundary element method. By applying the proposed method to character models, the effectiveness of the proposed model was confirmed.

    CiNii

  • Behavioral analysis of novice users on visual support.

    米村俊一, 吉田亨, 徳永幸生, 大谷淳

    電子情報通信学会技術研究報告   106 ( 84(HIP2006 1-22) )  2006

    J-GLOBAL

  • 自動会話システムにおける話題切り替えの効果

    河野恭佑, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   68th ( 2 )  2006

    J-GLOBAL

  • Behavioral analysis of novice users on visual support.

    米村俊一, 吉田亨, 徳永幸生, 大谷淳

    電子情報通信学会技術研究報告   106 ( 219(HCS2006 34-40) )  2006

    J-GLOBAL

  • 顔の輪郭画像を用いた映像通信システムにおける会話特性

    吉田亨, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   68th ( 2 )  2006

    J-GLOBAL

  • K-028 Proposal of Elastic Model for 3D Object with Articulated Structure

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

      4 ( 3 ) 417 - 418  2005.08

    CiNii

  • K-029 Space panel : a movement directing system for cyber-theater

    Kim Hyeonmoon, Ohya Jun, Ogata Hiroyuki

      4 ( 3 ) 419 - 420  2005.08

    CiNii

  • K-082 Studies of Black "●" Prompter Communication System

    Chen Li Jen, Ohya Jun, Tokunaga Yukio, Yonemura Shunichi

      4 ( 3 ) 543 - 544  2005.08

    CiNii

  • K-088 Cyber Actor's Motion Control Using Image Data

    Cho Jieun, Watanabe Takafumi, Ohya Jun

      4 ( 3 ) 559 - 560  2005.08

    CiNii

  • K-057 Study of Augmented Reality Based Method for Exhibiting Artistic Works with their Creation Processes

    Ton Srichompoo, Ohya Jun, Yonemura Syunichi, Enomoto Makoto

      4 ( 3 ) 485 - 486  2005.08

    CiNii

  • K-065 The Relationship between Caricatured Portrait and Dual Coding Theory in The Case of Face Recognition : Application to Caricatured Portrait Generate System Using Computer

    Enomoto Makoto, Ohya Jun, Kurumisawa Jun

      4 ( 3 ) 501 - 504  2005.08

    CiNii

  • I-061 Motion Detection Based On Active Stereo Camera Array

    Xie Yingdi, Ohya Jun

      4 ( 3 ) 145 - 146  2005.08

    CiNii

  • H-029 Interactive Visualization of Large ISO-Surfaces

    Sven Forstmann, Ohya Jun

      4 ( 2 ) 365 - 366  2005.08

     View Summary

    Presented is a novel algorithm for the real-time view-dependent visualization of large iso-surfaces. The developed method utilizes a unique continuous level-of-detail approach based on clip-boxes for visualizing the triangular geometry which is generated from the original iso-surface. The algorithm is especially focused on simplicity and performance for highly interactive and continuous frame rates.

    CiNii

  • L-055 A proposal on fire fighting support system using sensor networks and surveillance cameras

    Nogai Msahiro, Rubaiyat Yasmin, Ohya Jun

      4 ( 4 ) 133 - 134  2005.08

    CiNii

  • O-007 Basic Study for Recognition Method of Customers Behaviors Using Image Processing

    Hori Takayuki, Ohya Jun, Kurumisawa Jun

      4 ( 4 ) 391 - 392  2005.08

    CiNii

  • A Proposal on Active Extensible Stereo Camera Array For High Speed Moving Object Detection

    XIE Yingdi, OHYA Jun

    IPSJ SIG Notes   2005 ( 66 ) 43 - 48  2005.07

     View Summary

    Recent years, almost motion detection researches have been focused on detecting moving object with a low speed, while only a few have been considered high speed case. In this paper, a new approach for high speed moving object detection based on active extensible stereo camera array, and a novel approach of motion detection are presented. In our approach, the camera array is extensible by increasing or decreasing the number of the stereo camera set to scale the video sequence to a desired frame rate. In the experiment, this approach shows its ability of getting high frame rate images, which is the basic for high speed moving object detection.

    CiNii

  • A Proposal on Active Extensible Stereo Camera Array For High Speed Moving Object Detection

    XIE Yingdi, OHYA Jun

    IEICE technical report   105 ( 164 ) 43 - 48  2005.06

     View Summary

    Recent years, almost motion detection researches have been focused on detecting moving object with a low speed, while only a few have been considered high speed case. In this paper, a new approach for high speed moving object detection based on active extensible stereo camera array, and a novel approach of motion detection are presented. In our approach, the camera array is extensible by increasing or decreasing the number of the stereo camera set to scale the video sequence to a desired frame rate. In the experiment, this approach shows its ability of getting high frame rate images, which is the basic for high speed moving object detection.

    CiNii

  • A Painterly Rendering Method Based on the Soft/hard Edge and It's Evaluation

    SHINBARA Yusuke, TOKUNAGA Yukio, KURUMISAWA Jun, OHYA Jun

    IEICE technical report. Image engineering   104 ( 752 ) 1 - 6  2005.03

     View Summary

    In this paper, a method of computer graphics using the soft-edge and hard-edge techniques is proposed. These techniques give a reality to paintings because spaces among objects in paintings and solidity of objects can be described realistically. After artist's know-how is analyzed following some experiments, the soft-edge and hard-edge techniques are taken out as an algorithm. Opinion tests by young artists show usefulness of these techniques.

    CiNii

  • A Fundamental Analysis of Topic Changes Prompted by the "●" Communication System

    CHEN Li Jen, OHYA Jun, TOKUNAGA Yukio, KATO Tetsuya, YONEMURA Shunichi

    Technical report of IEICE. HCS   104 ( 744 ) 47 - 51  2005.03

     View Summary

    In this paper, data obtained from experiments into a new communication method using a black prompter, "●, " as a metaphor or talking partner, conducted using two different approaches, is analyzed. The experimental results showed that topic changes occurred more often on particular frames with similar patterns. The results have been examined and compared using several methodologies. Topic changes by the participants have been observed and studied.

    CiNii

  • Proposal of an Interactive Model for Articulated Elastic Objects

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

    Technical report of IEICE. Multimedia and virtual environment   104 ( 748 ) 7 - 12  2005.03

     View Summary

    This paper proposes an interactive model for animating complex-shaped articulated elastic objects. To each of a series of articulated bones, an elastic primitive is attached so that real-time animations based on the Boundary Element Method can be achieved. Inverse-kinematics is applied for computing the bones' transformations, and the lengths of the bones are changed so that the articulated elastic objects behave like one elastic object without any articulation. By enclosing the above-mentioned elastic object with an outer frame that consists of elastic primitives, multiple bones can be controlled simultaneously. Experimental results demonstrate interactive operations as well as realistic animations.

    CiNii

  • Fundamental Study of Estimating the Shape of the Surface of Water by Shape from Shading and Genetic Algorithms

    TERASHIMA Teppei, OHYA Jun, KURUMIZAWA Jun, TOKUNAGA Yukio

    Technical report of IEICE. Multimedia and virtual environment   104 ( 748 ) 1 - 6  2005.03

     View Summary

    The authors study to recognize water's behaviors by computer vision approach and reproduce the recognized behaviors using Bump Mapping in a virtual environment. To reproduce the shape of water's surfaces, it's necessary to estimate the orientations of surface normals to the surface. For this, we proposed to utilize Shape from Shading. However, since Shape from Shading assumes orthographic projection for the camera projection model, large estimation errors appear in pixels apart from the optical axis of the camera. To improve the estimation accuracy, one idea is to check the combination of all the possible surface normal's orientations, but this causes combinatorial explosion. This paper proposes a genetic algorithm based method for suppressing the errors efficiently, Experimental results using synthetic images show the effectiveness of the proposed method

    CiNii

  • A-16-23 A Study of the Action Timing of Cyber Actor

    Cho Jieun, Ohya Jun

    Proceedings of the IEICE General Conference   2005   299 - 299  2005.03

    CiNii

  • A-16-22 Theater Canvas : an interface instrument for Cyber-Theater directing system

    Kim Hyeonmoon, Cho Jieun, Ohya Jun

    Proceedings of the IEICE General Conference   2005   298 - 298  2005.03

    CiNii

  • A-16-10 Study of Expressing Avatars for Cyber Theater : Improving Computer Vision-based Human Posture Estimation by Neural Networks

    Iida Masaya, Kang Dong Wan, Ohya Jun

    Proceedings of the IEICE General Conference   2005   286 - 286  2005.03

    CiNii

  • D-12-130 A Study of Improving the Estimation Accuracy of the Position and Orientation of an Augmented Reality Terminal Using Factorization Method

    Fujiwara Yuya, Ohya Jun, Nogai Mashiro

    Proceedings of the IEICE General Conference   2005 ( 2 ) 280 - 280  2005.03

    CiNii

  • D-12-79 A Study of Synthesizing New Human Motions from Sampled Motions Using Tensor Decomposition

    Rovshan Kalanov, Cho Jieun, Ohya Jun

    Proceedings of the IEICE General Conference   2005 ( 2 ) 229 - 229  2005.03

    CiNii

  • D-12-77 Study of Articulated Model for Elastic Objects by Inverse Kinematics

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

    Proceedings of the IEICE General Conference   2005 ( 2 ) 227 - 227  2005.03

    CiNii

  • D-12-11 Study of Estimating Three Dimensional Postures of a Human Using Color Information Processing

    Onuma Yuichi, Kang Dong Wan, Ohya Jun

    Proceedings of the IEICE General Conference   2005 ( 2 ) 161 - 161  2005.03

    CiNii

  • D-12-4 Fundamental Study of Estimating the Shape of the Surface of Water by Shape from Shading and Genetic Algorithms

    Terashima Teppei, Ohya Jun, Kurumizawa Jun, Tokunaga Kousei

    Proceedings of the IEICE General Conference   2005 ( 2 ) 154 - 154  2005.03

    CiNii

  • D-4-6 Analysis-based Participant Selection Scheme in Workflow Management System

    Ding Xiao, Cao Jian, Ohya Jun

    Proceedings of the IEICE General Conference   2005 ( 1 ) 32 - 32  2005.03

    CiNii

  • 顔のエッジ表現を用いたコミュニケーションシステムの会話特性

    大石貴也, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会全国大会講演論文集   67th ( 4 )  2005

    J-GLOBAL

  • Visual Communication System Based on Edge Abstraction

    大石貴也, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会研究報告   2005 ( 9(HI-112) )  2005

    J-GLOBAL

  • 熱転写記録を用いた金色銀色表現法とその用応

    徳永 幸生, 古田 裕樹, 楜沢 順, 大谷 淳

    パソコンリテラシ   29 ( 10 ) 19 - 23  2004.10

    CiNii

  • K-069 Study on Manipulation Method for Animating Elastic Objects

    Watanabe Takafumi, Ohya Jun

      3 ( 3 ) 549 - 550  2004.08

    CiNii

  • K-044 A Communication System using a black prompter "●"

    Chen Li Jen, Kato Tetsuya, Tokunaga Yukio, Ohya Jun

      3 ( 3 ) 491 - 492  2004.08

    CiNii

  • I-053 Study of estimating surface normals to water surfaces using Shape from Shading and Affine transformation for the original image

    Terashima Teppei, Ohya Jun

      3 ( 3 ) 123 - 124  2004.08

    CiNii

  • K-032 A Study of Improving the Estimation Accuracy of the Position and Orientation of an Augmented Reality Terminal in a Wide Area

    Fujiwara Yuya, Ohya Jun

      3 ( 3 ) 467 - 468  2004.08

    CiNii

  • K-033 Study of Estimating Three Dimensional Postures of a Human Using Color Information Processing and Time Series Information Processing

    Onuma Yuichi, Kang Dong-Wan, Ohya Jun

      3 ( 3 ) 469 - 470  2004.08

    CiNii

  • K-035 A Frame Work for Analysis and Synthesis of Actor's Stylistic Action

    Cho Jieun, Ohya Jun

      3 ( 3 ) 473 - 474  2004.08

    CiNii

  • Preliminary Study on Auditory and Vision Interaction with Contrast Stimuli

    KOBAYASHI Yuichi, SHIBATA Tatsuya, KATO Toshikazu, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   103 ( 745 ) 7 - 12  2004.03

     View Summary

    We have been studying human subjective impressions for various visual contents. Substantially, people have an impression when not only the visual but also the other senses are activated. In this report, we have focused on the relationship between the vision and the auditory. In order to investigate their mutual affects, we focused on contrast. In this report, our purpose is to grasp the stimulus range where the interaction between the auditory and the vision can be seen. We performed psychological experiments with visual contrast stimuli and audio contrast stimuli. Accordingly, we found that the tendency of vision dominance over auditory can be seen when human sensitivity to luminance is bad such as in the case with brighter or darker stimuli. Moreover, the intensity of the tendency is stronger for inharmonic auditory stimuli than for harmonic auditory stimuli. As a result of the ANOVA test, the main effects are stronger when both visual stimuli and audio stimuli increase or decrease and the interaction is stronger when visual stimuli increases and audio stimuli decreases or vice versa. These results suggest that cross-modal effects are determined by the reliability of the visual and auditory information.

    CiNii

  • Study of a Digital Method for Displaying Medieval Tempera Paintings : Attempt to Restoring Original Colors and Preserving Sensations of Materials

    SAITO Minoru, OHYA Jun, KURUMISAWA Jun, HIRAO Daisuke, LU Liang Chen, TOKUNAGA Yukio

    IEICE technical report. Image engineering   103 ( 739 ) 1 - 6  2004.03

     View Summary

    In most of currently existing digital library systems, the artistic works stored in the systems are displayed in personal computers' monitors, but the original colors are not restored, and the sensations of materials, which the artistic works present, are lost. Oil paintings and water colors are typical examples of artistic paintings, but it is very difficult to analyze these paintings, because the former utilizes glazing, in which pigments with transparencies are overlaid; the latter utilizes gradation, which is achieved by precise mixture of water and pigments. Therefore, this paper deals with early Medieval Tempera paintings, in which only opaque pigments were used. This paper reports a fundamental study of a digital display method that restores the original colors and preserves the sensations of materials. An sRGB monitor based system that displays the original colors of the pigments used for the Tempera paintings is constructed based on subjective tests in which the original and displayed colors of the pigments are compared. At each pixel of a digital Tempera painting image, the pigment used for that pixel is estimated by the Nearest Neighbor rule; then, the color of that pigment is displayed to the sRGB monitor. Experiments using 3 works stored in the Uffizi Gallery show some promising results.

    CiNii

  • Study of Estimating the Direction and Strength of Wind by Analyzing Botanical Trees' Behaviors in Video Sequences for Reproducing the Behaviors in Virtual Space

    LU LiangChen, OHYA Jun, TERASHIMA Teppei

    Technical report of IEICE. PRMU   103 ( 737 ) 103 - 108  2004.03

     View Summary

    This paper deals with a method that analyzes a botanical tree's behaviors in real space by a computer vision approach so as to reproduce the analyzed behaviors in virtual space. Instead of applying unstable local tracking to the tree in a video sequence, we estimate the direction and strength of the wind that shakes the tree by a learning based method that classifies the input video sequence into one of the stored winds with different directions and strengths. In the learning phase, sample video sequences are used for constructing the Eigenspace and Fisherspace, which is obtained from Fisher discriminant analysis. In the classification phase, the input video sequence is compared with each of the stored sample sequences so that the direction and strength of the wind are estimated. An interplocation method improves the estimation accuracy. Experimental results demonstrate the effectiveness of the proposed method.

    CiNii

  • D-11-81 Study of Restoring Original Colors of Digital Images of Medieval Tempera Painting

    Saito Minoru, Ohya Jun, Kurumisawa Jun, Hirao Daisuke, Chen Lu Liang, Tokunaga Yukio

    Proceedings of the IEICE General Conference   2004 ( 2 ) 81 - 81  2004.03

    CiNii

  • D-11-120 Study of Real-time Animation of 3D Elastic Objects by the Boundary Element Method Using Vertex Interpolations

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

    Proceedings of the IEICE General Conference   2004 ( 2 ) 120 - 120  2004.03

    CiNii

  • A-15-19 Preliminary Study on Audio Visual Interaction

    Kobayashi Yuichi, Shibata Tatsuya, Kato Toshikazu, Ohya Jun

    Proceedings of the IEICE General Conference   2004   299 - 299  2004.03

    CiNii

  • A-16-4 A Study of Improving the Estimation Accuracy of the Position and Orientation of an Augmented Reality Terminal Using Infrared LED's

    Fujiwara Yuya, Ohya Jun

    Proceedings of the IEICE General Conference   2004   313 - 313  2004.03

    CiNii

  • A-16-19 A 3D Facial Image Reconstruction Method Using Shape Deformation and Texture Remapping

    Sato Hajime, Ohya Jun, Terashima Nobuyoshi

    Proceedings of the IEICE General Conference   2004   328 - 328  2004.03

    CiNii

  • D-12-116 Estimating Complicated Posture of a Human Wearing a Multiple-Colored Suit Using Color Information Processing

    Kang Dong Wan, Onuma Yuichi, Ohya Jun

    Proceedings of the IEICE General Conference   2004 ( 2 ) 282 - 282  2004.03

    CiNii

  • Gold and Silver Imaging Method by Thermal Transfer Printer, and Its Application

    FURUTA Hiroki, TOKUNAGA Yukio, KURUMISAWA Jun, OHYA Jun

    ITE technical report   28 ( 9 ) 25 - 28  2004.02

    CiNii

  • Gold and Silver Imaging Method by Thermal Transfer Printer, and Its Application

    FURUTA Hiroki, TOKUNAGA Yukio, KURUMISAWA Jun, OHYA Jun

    IEICE technical report. Electronic information displays   103 ( 649 ) 25 - 28  2004.02

     View Summary

    Gold/silver imaging method is developed, including digital color conversion table of CMYKGS and CMYK(GS), standard 216 CMY colors map on gold/silver layer and printing procedure for the CMY color using Thermal Transfer Printer. Digital printing arts and new digital expression of "Kin Byoubu" are printed out with sufficiently good performance level.

    CiNii

  • Real-Time Stereo by using Dynamic Programming

    FORSTMANN Sven, KANOU Yutaka, OHYA Jun, THUERING Sven, SCHMITT Alfred

    Technical report of IEICE. HIP   103 ( 455 ) 103 - 108  2003.11

     View Summary

    A dynamic programming (DP) based algorithm that achieves real-time, high quality stereo-matching is presented. A special coarse to fine approach as well as the MMX extensions based hardwares increase the computation speed. Vertical smoothing and integration of the path in the DP matching make it possible to avoid noisy horizontal strokes so that high quality stereo-matching is achieved. The current implementation runs at about 30 FPS on a 2GHz PC, which is enough to be used for real-time applications.

    CiNii

  • Real-Time Animation Model for Non-Constraint Elastic Objects : Boundary Element Method Model and Reproduction of Actual Behaviors

    WATANABE Takafumi, OHYA Jun, KURUMISAWA Jun, TOKUNAGA Yukio

    Technical report of IEICE. Multimedia and virtual environment   103 ( 352 ) 35 - 40  2003.10

     View Summary

    We present a new real-time method that can animate a moving elastic object that could collide with other objects in a virtual environment. For simulating physical motions of elastic objects, we exploit a Boundary Element Method (BEM), which can achieve efficient computations and desired deformation as opposed to spring models and a Finite Element Method. In case that real-time processes are required for the BEM, only the physical simulation cannot achieve animations; therefore, this paper proposes a 2D model that is combined with the BEM. Since it is difficult to animate 3D elastic objects in real-time, we propose a method that approximates the 3D motions based on the 2D model. The effectiveness of the proposed method is demonstrated by the experiments in which the dynamical behaviors of a jelly are reproduced in a virtual environment in real-time.

    CiNii

  • Study of Emotion-based Variations in Performers' Body Motions in the Cyber Theater

    CHO Jieun, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   103 ( 351 ) 31 - 36  2003.10

     View Summary

    Towards the realization of the Cyber Theater, which enables anybody to produce contents as a performer and director and to see the contents as an audience in the virtual environment generated in the networks, this paper proposes a new directing system for the Cyber Theater and conducts fundamental studies. The goal of the proposed directing system is to make it possible for the director to produce the performance scenes the director desires by editing the performance fragments stored in the database. That is, the goal is to produce variable actions of the cyber characters by editing the fragments stored in the database. We experimentally confirmed that the director can edit cyber characters' performances by giving emotion-based variations to each action. It turned out to us that multiple directions can be achieved by adjusting the strength of the emotion-based actions. These results show a possibility that new directions, which are different from directions in real space, can be realized in the Cyber Theater.

    CiNii

  • Construction of 3D Human Face Images using Video Images and a Facial Muscle Model

    Sato Hajime, Ohya Jun, Terashima Nobuyoshi

    Proceedings of the IEICE General Conference   2003 ( 2 ) 325 - 325  2003.03

    CiNii

  • Real Time Reproduction of 3D Jelly like objects' Behaviors by the Boundary Element Method

    Watanabe Takahumi, Ohya Jun

    Proceedings of the IEICE General Conference   2003 ( 2 ) 258 - 258  2003.03

    CiNii

  • The Cyber-Theater Directing System based on the Analysis of Drama Scripts

    Cho Jieun, Ohya Jun

    Proceedings of the IEICE General Conference   2003   326 - 326  2003.03

    CiNii

  • Study of Efficient, Realistic Animation of 3D Botanical Tree Model

    KANDA Hitoshi, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   102 ( 662 ) 81 - 86  2003.02

     View Summary

    This paper proposes a new efficient method that can animate botanical trees in 3D realistically. In this paper, a 3D botanical tree model consists of a set of branch segments, to which leaf models are attached. To reduce the amount of computation, instead of calculating the motions of all the branch segments, only the representative segment in each branch is numerically analyzed. The numerical analysis is constrained to a 2D plane so that 3D numerical analysis need not be performed. Concerning the leaf model, a set of some leaves is systematically attached to each branch segment. Experimental results clarify the conditions for real-time, realistic animations of dynamic behaviors of trees.

    CiNii

  • Estimating the Surface Shape from Shading of Water Based on Shape from Shading and Its Reproduction in Virtual Environments

    TERASHIMA Teppei, OHYA Jun

    Technical report of IEICE. PRMU   102 ( 652 ) 55 - 60  2003.02

    CiNii

  • Generation of 3D Facial Images Using Video Images and a Facial Muscle Based Model

    SATO Hajime, OHYA Jun, TERASHIMA Nobuyoshi

    Technical report of IEICE. PRMU   102 ( 651 ) 53 - 56  2003.02

     View Summary

    A ligh speed, high precision method for constructing 3D human facial images utilizing the characteristics of video images directly is proposed. A 3D facial model, composed of 3D polygons and a 2D facial muscle model, is constructed from two orthogonal facial images. Video input frames are modified, and the movements of facial feature points are measured based on the the contraction rule of facial muscles. The vertices of the 3D polygons are altered adaptively according to the positions of the feature points. and the modified input frame is texture mapped directly onto to the model. This method was implemented using an ordinary CCD camera and computer, and a fundamental experiment using one person's data showed its effectiveness.

    CiNii

  • Research in the ubiquitous archives in the field of cultural properties. -About ubiquitous archives which take on RFID (Redio Frequency Identification) and IC tip/tag, in the cultural properties.-

    KOIKE Koji, HIRAO Daisuke, OHYA Jun

    IPSJ SIG Notes   2002 ( 107 ) 47 - 54  2002.11

     View Summary

    The preservation, management and learning system to use the short-range wireless cellular phone and the ultraminiature IC chip loading the antenna in the field of cultural properties. Putting the chip on various objects makes it a new, ubiquitous, and distributed database, which can directly get the needed information.

    CiNii

  • J-41 A Real-time Method for Animating Virtual Elastic Objects by Use of a Boundary Element Method

    Watanabe Takahumi, Ohya Jun

      2002 ( 3 ) 283 - 284  2002.09

    CiNii

  • J-11 Study of 3D Botanical Tree Modeling That Enables Efficient Animations of Dynamic Behaviors

    Kanda Hitoshi, Ohya Jun

      2002 ( 3 ) 223 - 224  2002.09

    CiNii

  • I-72 Study of Image Processing Based Recognition of Ancient Coins Illuminated from Three Directions

    Hashimoto Jun, Ohya Jun

      2002 ( 3 ) 143 - 144  2002.09

    CiNii

  • K-30 Study of measuring the absolute position and orientation for achieving a Field Augmented Reality.

    Sakamoto Atsuya, Koike Koji, Ohya Jun

      2002 ( 3 ) 427 - 428  2002.09

    CiNii

  • K-20 Material Analysis for 3D texture images

    Kobayahsi Yuichi, Ohya Jun, Tadenuma Makoto

      2002 ( 3 ) 407 - 408  2002.09

    CiNii

  • I-22 Study of estimating postures of a human wearing a multiple-colored suit based on color information processing

    Kang Dong-Wan, Ohya Jun

      2002 ( 3 ) 43 - 44  2002.09

    CiNii

  • Construction of 3D Facial Feature Models Considering Facial Muscle Movements Using Two Photographs

    SATO Hajime, LIU Yi-Chih, OHYA Jun, TERASHIMA Nobuyoshi

    IEICE technical report. Image engineering   102 ( 216 ) 77 - 81  2002.07

     View Summary

    A method for constructing facial images from arbitrary directions and expressions for use in multimedia communication applications is proposed. Both the geometric deformations and physical movements of the facial muscles are considered in this method, and enables the construction of realistic facial images in a simple way. Two orthogonal images of an expressionless face are acquired, and feature points are selected to construct a three dimensional polygon model to describe the shape of the human face. Elastic springs are placed on the model to express the deformations of facial components, and polygon vertices near the springs' dynamic points are displaced accordingly to construct changes in expressions naturally. Since these spring movements do not depend on a specific facial structure, the model could easily be adapted to any given person's face. The method was implemented on an inexpensive PC, and simulation results showed its effectiveness.

    CiNii

  • Construction of Muscle-Based Face Model and Its Application to Facial Expression Synthesis

    Liu Yi-Chih, Sato Hajime, Ohya Jun

    Proceedings of the IEICE General Conference   2002   355 - 355  2002.03

    CiNii

  • Study of reproducing fluid's behaviors in a virtual environment using a video seguenc

    Tarashima Teppei, Ohya Jun

    Proceedings of the Society Conference of IEICE   2001   161 - 161  2001.08

    CiNii

  • Visual Grouping and Composition in Art-Based Rendering

    KOWALSKI Michael A., HUGHES John F., RUBIN Cynthia B., OHYA Jun

    Technical report of IEICE. DSP   100 ( 12 ) 13 - 18  2000.04

     View Summary

    We apply techniques from traditional artistic composition to the art-based rendering of interactive 3D scenes. Our system dynamically adjusts the rendering attributes of objects in the scene to achieve the desired compositional effects for a given view. One such effect is that of selectively grouping scene elements through shared tone, color, and outline. We demonstrate how such grouping can simplify and structure the image, and how it can be used, together with controlled level of detail, to emphasize important objects in the scene. We also show how existing shading, silhouette-extraction, and detail-generation algorithms can be adapted to support such compositional effects.

    CiNii

  • Analysis of Movements of Facial Expressions by High Speed Camera

    UCHIDA Hideko, YOTSUKURA Tatsuo, MORISHIMA Shigeo, YAMADA Hiroshi, OHYA Jun, AKAMATSU Shigeru

    Technical report of IEICE. HIP   99 ( 722 ) 1 - 6  2000.03

     View Summary

    The purpose of this study was to examine patterns of facial movements of "posed" (intended) facial expressions and "elicited" (unintended) emotional responding by feature point tracking. We videotaped participants' facial movements of intended and unintended facial expressions of emotion by a high speed camera which allowed us to analyze facial movements very closely in image sequences. The experiment consisted of two parts. First, the participants task was to produce the six expressions (anger, disgust, fear, happiness, sadness, and surprise). In the second part of the experiment, participants were shown a set of film stimuli that elicited each of emotional states (amusemet, anger, disgust, fear, sadness, and surprise). We recorded the participants' facial expressions to the film stimuli.

    CiNii

  • Analysis of Facial Behevior by Using High-Speed Camera

    Yotsukura Tatsuo, Uchida Hideko, Yamada Hiroshi, Morishima Shigeo, Akamatsu Shigeru, Ohya Jun

    Proceedings of the IEICE General Conference   2000   260 - 260  2000.03

    CiNii

  • Research on Virtual Communication Environments at ATR

    OHYA Jun, SAKAGUCHI Tatsumi

    Technical report of IEICE. PRMU   99 ( 574 ) 79 - 84  2000.01

     View Summary

    This article describes ATR's research activities on virtual communication environments in which remotely located people can communicate with each other through a virtual scene. To realize such environments, human image analysis and synthesis and scene image analysis and synthesis are essential. Concerning to human image analysis and synthesis, the virtual metamorphosis systems and hand gesture recognition from multiple camera images are introduced. Regarding scene image analysis and synthesis, novel view generation from real images, real image based 3D modeling that preserves the reality of appearance and behaiors, and merging real world and CG images are described. Our on-going projects are also summarize.

    CiNii

  • Remarks on Real-time Estimation of Human Body Postures using Neural Networks

    Uemura Tetsuya, Takahashi Kazuhiko, Ohya Jun

    Proceedings of the IEICE General Conference     302 - 302  2000

    CiNii

  • Stabilization of Multiple Human Tracking Using Non-synchronous Multiple Viewpoint Observations

    YANG Howard Y, UTSUMI Akira, OHYA Jun

    Technical report of IEICE. PRMU   99 ( 450 ) 1 - 7  1999.11

     View Summary

    We propose a method to track multiple human motion using multiple-camera-based non-synchronous observations. Our method employed Kalman-filtering algorithm to integrate non-synchronous observations from multiple viewpoints effectively. Our experimental system consists of three types of processes, discovering node for detecting newly appeared person, tracking node for tracking each target person, and observation node for processing each viewpoint images. In observation node, occlusion occurred among multiple persons cause serious problem to the matching process between the observed feature and tracked model. In the proposed method, not only position but also human features (height and cloth color) are employed for the matching process to enhance the tracking accuracy. Experimental results with real scene images support our system effectiveness.

    CiNii

  • "Bunshin" Communication : New Communication Means Utilizing Non-verbal Information

    OHYA Jun, NAKATSU Ryohei, KAWATO Shinjiro, SAKAGUCHI Tatsumi

    Technical report of IEICE. PRMU   99 ( 448 ) 47 - 52  1999.11

     View Summary

    This paper proposes the "Bunshin" communication, which enables a person to participate in simultaneous multiple events such as meetings and lectures as a virtual-metamorphosed avatar and/or agents. Virtual metamorphosis and agents are switched according to the recognized non-verbal data generated by the other participants at each event. Bunshin communication can achieve efficient use of time and daily life full of humanity and makes it possible to transmit the person's opinions and collect necessary information. The effectiveness of the Bunshin communication is confirmed by the prototype system we recently built.

    CiNii

  • Eye and Lip Detection and Tracking Using Active Camera

    YOTSUKURA Tatsuo, SHIMADA Naoyuki, MORISHIMA Shigeo, OHYA Jun

    Technical report of IEICE. HIP   99 ( 451 ) 31 - 36  1999.11

     View Summary

    We propose a technique to track user's eye and mouth by using two Pan-Tilt-Zoom controllable cameras. Mouth and eye zones are detected by combine the binary images from the camera. The zoom, rotation direction, and capturing rate of the cameras are automatically changed depends on the results of the captured images. By studying the characteristics of the extracted binary images, we are able to track lip and eye movements such as Lip Reading and blink. Experiments using the technique have shown satisfactory result.

    CiNii

  • Real-time, Non-contact, No-wear, 3D Human Body Posture Estimation Method using Trinocular Images

    TAKAHASHI Kazuhiko, SAKAGUCHI Tatsumi, OHYA Jyun

    Technical Report of IEICE   99 ( 380 ) 47 - 54  1999.10

     View Summary

    This paper proposes a real-time method for estimating human postures in 3D from camera images. In this method, the feature points of the human body (top of the head, and tips of hands and feet) are obtained from either the results of human contour analysis or those of time subtraction images. To optimize their positions and to overcome self-occlusion problems, all feature points are tracked by the AR model using the Kalman filter. The proposed method is applied to the trinocular camera systems and 3D human body postures are estimated. The entire process runs in real-time on a personal computer. Experimental results show the feasibility and effectiveness of the proposed method.

    CiNii

  • 多数カメラによる両手手振り認識を用いた仮想空間操作

    内海章, 大谷淳

    ヒューマンインタフェースシンポジウム論文集   1999   227 - 232  1999.10

    J-GLOBAL

  • Face Location Detection: a Step to Facial Expression Estimation

    KAWATO Shinjiro, OHYA Jun

    Proceedings of the Society Conference of IEICE   1999   203 - 203  1999.08

    CiNii

  • Remarks on Real-time Estimation of Human Body Posture using Kalman Filter

    Takahashi Kazuhiko, Sakaguchi Tatsumi, Ohya Jun

    Proceedings of the Society Conference of IEICE   1999   174 - 174  1999.08

    CiNii

  • 非同期多視点画像を用いた人物追跡システムの構築

    内海章, 大谷淳

    電子情報通信学会大会講演論文集   1999   173  1999.08

    J-GLOBAL

  • Human Motion Detection using Multiple Camera System

    MORI Hiroki, UTSUMI Akira, OHYA Jun, YACHIDA Masahiko

    Technical report of IEICE. Multimedia and virtual environment   98 ( 684(MVE98 98-104) ) 41 - 48  1999.03

     View Summary

    We propose a method to detect human motion information using multiple viewpoint system. In vision-based human tracking, self-occlusions and human-human occlusions are a part of the more significant problems. We employ multiple-viewpoint system to reduce these problems. Our system can retrieve both position and orientation of human bodies, face images and hand positions. Experimental results with five cameras are also described.

    CiNii J-GLOBAL

  • Dynamics Animation of Botanical Tree for Interactive Environment

    Sakaguchi Tatsumi, Ohya Jun

    Proceedings of the IEICE General Conference   1999   370 - 370  1999.03

    CiNii

  • Study on Extraction of Atomic Facial Motion

    Otsuka Takahiro, Ohya Jun

    Proceedings of the IEICE General Conference   1999 ( 2 ) 254 - 254  1999.03

    CiNii

  • Realtime facial expression estimation adapted to head rotations and translations

    TAKEMATSU Katsuhiro, EBIHARA Kazuyuki, SAKAGUCHI Tatsumi, OHYA Jun

    Proceedings of the IEICE General Conference   1999 ( 2 ) 253 - 253  1999.03

    CiNii

  • Visual Feedback Device for Non-contact Gesture Interaction in Real Environment

    UTSUMI Akira, OHYA Jun

    Proceedings of the IEICE General Conference   1999   357 - 357  1999.03

    CiNii

  • Human Body Posture Estimation from Trinocular Silhouette Images

    Iwasawa Shoichiro, Ohya Jun, Morishima Shigeo

    Proceedings of the IEICE General Conference   1999 ( 2 ) 264 - 264  1999.03

    CiNii

  • A Study of a Non-Verbal Information Transfer Method for a Communication Environment : Head Pose Estimation, Facial Expression Recognition and Conversion

    OTSUKA Takahiro, OHYA Jun

    Technical report of IEICE. PRMU   98 ( 608 ) 33 - 39  1999.02

     View Summary

    A method is developed that can enhance the transfer of non-verbal information in a communication environment by rendering 3D model of the user's face in which the position and degree of deformation is controlled by the recognition result of facial expressions. In this method, first, the head movement and the muscle motions are estimated by tracking the feature points in the segmented head region. Then, facial expression is recognized by comparing the muscle contraction vector to the probabilistic distribution for the six fundamental expressions. Finally, the face animation is rendered such that the corresponding muscles for the recognized expression are activated. Experimental results showed the accurate detection of muscle contraction for facial expression.

    CiNii

  • Multiple-Human Tracking using Asynchronous Multiple Cameras

    MORI Hiroki, UTSUMI Akira, OHYA Jun, YACHIDA Masahiko

    Technical report of IEICE. PRMU   98 ( 527 ) 15 - 22  1999.01

     View Summary

    We propose a human motion tracking method using multiple-viewpoint images. Multiple-view-based approach is effective for tracking wide range human motions. However, conventional tracking systems require a camera synchronization mechanism and it restricts its scalability. In the system, the slowest image processing determines the entire speed. To overcome these problems, we propose an asynchronous system to track multiple person. In our system, each camera image can be processed independently with a PC. Observations in multiple viewpoints are integrated with Kalman filtering algorithm. Results of some fundamental experiments are described.

    CiNii

  • 2)P型フーリエ記述子を用いた多数カメラによる手形状認識(マルチメディア情報処理研究会 ネットワーク映像メディア研究会合同)

    内海 章, 大谷 淳

    映像情報メディア学会誌 : 映像情報メディア   52 ( 12 )  1998.12

    CiNii

  • Multiple Camera Image Processing based Real-time Motion Capture for "Shall We Dance?" Project

    IWASAWA SHOICHIRO, EBIHARA KAZUYUKI, TAKEMATSU KATSUHIRO, SAKAGUCHI TATSUMI, OHYA JUN

    Technical report of IEICE. PRMU   98 ( 394 ) 15 - 22  1998.11

     View Summary

    This report describes a virtual metamorphosis system called "Shall We Dance?" which includes a real-time 3D computer vision system for detecting and estimating human body posture and providing a person with control over the movement of a virtual puppet. Previously, the authors proposed the 2D posture estimation method only uses a monocular image. The new method acquires three observation view of human body and the body parts are detected and located in real-time basis done by a heuristic analysis of each image. Then 2D locations of body parts are evaluated and an optimal pair of images are used for recovering 3D body posture. Experimental results show the usefulness of the proposed method and real-time performance.

    CiNii

  • Study on Real-time Estimation of Head Motion Using Weak Perspective Epipolar Geometry

    Otsuka Takahiro, Ohya Jun

    Proceedings of the Society Conference of IEICE   1998   306 - 306  1998.09

    CiNii

  • Kalman-Filter-Based Human Tracking using Asynchronous Multiple Cameras

    UTSUMI Akira, OHYA Jun

    Proceedings of the Society Conference of IEICE   1998   274 - 274  1998.09

    CiNii J-GLOBAL

  • Real-Time Human Body Posture Estimation from Multiple Images

    Iwasawa Shoichiro, Takematsu Katsuhiro, Ohya Jun, Morishima Shigeo

    Proceedings of the Society Conference of IEICE   1998   308 - 308  1998.09

    CiNii

  • Multiple-Camera-Based Hand Gesture Recognition using P-Type Fourier Descriptor

    Utsumi Akira, Ohya Jun

    ITE Technical Report   22 ( 39(MIP98 56-63/NIM98 82-89) ) 7 - 12  1998.07

     View Summary

    We propose a hand shape recognition system using multiple-viewpoint images. Most conventional hand shape recognition system did not concern about self-occlusion problem due to the hand rotation and a user needs to pay attention to his hand's direction to avoid the problem. We employ multiple-viewpoint images to estimate the pose of a human hand. After a pose estimation, a "best view" for the hand shape recognition is selected based on the estimation result. Hand shape recognition is performed based on the shape representation using P-type Fourier descriptor that is not affected with image scaling and translation. Based on the shape recognition, we developed a system where a user co create virtual graphical scenes interactively. In the system, a user can change the virtual objects' positions, sizes, colors, etc. with hand gestures. This system can be used as a user interface device, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices.

    DOI CiNii J-GLOBAL

  • Botanical tree modeling based on real images

    Sakaguchi Tatsumi, Ohya Jun

    Proceedings of the IEICE General Conference   1998 ( 2 ) 128 - 128  1998.03

    CiNii

  • Real time facial expression detection and reproduction using active camera system

    Takematsu Katsuhiro, Ebihara Kazuyuki, Ohya Jun

    Proceedings of the IEICE General Conference   1998 ( 2 ) 340 - 340  1998.03

    CiNii

  • Study of Spotting and Recognition of Facial Expressions from Image Sequences with Cancelling Head Motions

    Otsuka Takahiro, Ohya Jun

    Proceedings of the IEICE General Conference   1998 ( 2 ) 341 - 341  1998.03

    CiNii

  • Painting Process in Another BRAIN

    Kurumisawa Jun, Ohya Jun

    Proceedings of the IEICE General Conference   1998   373 - 373  1998.03

    CiNii

  • Multiple-Camera-Based Multiple-Human Tracking using Kalman Filtering

    UTSUMI Akira, MORI Hiroki, OHYA Jun, YACHIDA Masahiko

    Proceedings of the IEICE General Conference   1998 ( 2 ) 345 - 345  1998.03

    CiNii J-GLOBAL

  • An Active Space Indexing System for 3D Estimation of Human Postures

    SEMWAL Sudhanshu, OHYA Jun, FERMIN Iris

    IPSJ SIG Notes. CVIM   1998 ( 5 ) 89 - 96  1998.01

     View Summary

    We present a method for unencumbered tracking participants in a virtual environment using multiple cameras. The method consists of two main, modules: preprocessing and 3D posture estimation. In the preprocessing step the 3D space (we call active space) is indexed in advance, and in the estimation step, the 3D positions of some significant points of the body of a participant, are obtained based on that indexed space.

    CiNii

  • Real time facial expression detection using computer vision

    TAKEMATSU Katsuhiro, EBIHARA Kazuyuki, OHYA Jun

    Technical report of IEICE. PRMU   97 ( 500 ) 63 - 68  1998.01

     View Summary

    The authors are trying to realize a virtual communication environment in which remotely located people can communicate with each other via a virtual scene in a variety of fashions. To realize such an environment, facial expression estimation is a very important technology. In the authors' conventional system a person needs to wear a helmet to which a CCD camera is fixed. In this paper, we propose a new non-contact type facial expression system in which an active camera is used to track a human head using skin color information of the face. Experimental results show the effectiveness of the proposed method.

    CiNii

  • Research Activities on the Estimation of Facial Expressions and Body Postures in ATR Media Integration & Communications Research Laboratories

    OHYA Jun

    Technical report of IEICE. PRMU   97 ( 500 ) 77 - 81  1998.01

     View Summary

    At ATR Media Integration &amp Communications Research Labs., the authors are trying to realize virtual communication environments in which remotely located people can communicate with each other via a virtual scene in a variety of fashions. A virtual metamorphosis system, in which people can change their forms to any other forms, is an example of such an environment. To realize the virtual metamorphosis system, it is necessary to estimate facial expressions and body postures of a person and to reproduce them in the form to which the person changes his/her form. This report describes the authors' methods for estimating facial expressions and body postures as well as research works on facial expression recognition and hand posture recognition.

    CiNii

  • 多数カメラによる複数人物の位置・姿勢推定手法の検討

    森大樹, 内海章, 大谷淳, 谷内田正彦

    電子情報通信学会技術研究報告   97 ( 500(PRMU97 196-206) ) 23 - 30  1998.01

    J-GLOBAL

  • Dynamics Animation of Swaying Branches by Wind

    Sakaguchi Tatsumi, Ohya Jun

    Proceedings of the Society Conference of IEICE     192 - 192  1998

    CiNii

  • ATR知能映像通信研究所

    大谷 淳, 間瀬 健二, 宮里 勉

    日本バーチャルリアリティ学会誌 = Journal of the Virtual Reality Society of Japan   2 ( 2 ) 110 - 112  1997.12

    CiNii

  • Botanical Tree Modeling Based on Multiple Images

    SAKAGUCHI Tatsumi, OHYA Jun

    Technical report of IEICE. Multimedia and virtual environment   97 ( 449 ) 25 - 32  1997.12

     View Summary

    To generate virtual scenes through which remotely located people communicate with each other, it is very important to create realistic virtual scenes with which the people can interact. This paper proposes a new automatic method for modeling botanical trees as an example of objects in natural scenes, based on real images. The silhouette of the tree in multiple images are extracted from the background so that 3D volume data for the multiple images are constructed. The volume data are used to generate branches of the tree according to the rules based on a growth model of trees. After the generation of the branches, leaf models with color texture are attached to the branches. Some rendering results demostrate the effectiveness of the proposed method.

    CiNii

  • Real-time Estimation of Human Body Postures for PC Base MICS

    YAMADA Masanori, EBIHARA Kazuyuki, OHYA Jun

    ITE Winter Annual Convention   ( 1997 ) 104 - 104  1997.12

    CiNii

  • A Study of Spotting Segments Displaying Facial Expression from Image Sequences Using HMM

    OTSUKA Takahiro, OHYA Jun

    PRMU   97 ( 387 ) 17 - 24  1997.11

     View Summary

    A method that can be used for spotting segments that display facial expression is proposed. The motion of the face is modeled by HMM in such a way that each state corresponds to the conditions of facial muscles, e.g., relaxed, contracting, apex and relaxing. The probability assigned to each state is updated iteratively as the feature vector is obtained from image processing. A spotted segment is placed into a certain category when the probability of that category exceeds a threshold value. Experiments showed that the segments for the six basic expressions have been spotted accurately in near real time.

    CiNii

  • 11)実時間人物全身像動き検出手法([マルチメディア情報処理研究会 ネットワーク映像メディア研究会]合同)

    山田 正紀, 海老原 一之, 大谷 淳

    映像情報メディア学会誌 : 映像情報メディア   51 ( 11 ) 1859 - 1859  1997.11

    CiNii

  • 5)多数カメラによる人物位置・姿勢推定(〔ヒューマンインフォメーション研究会 ネットワーク映像メディア研究会〕合同)

    森 大樹, 谷内田 正彦, 内海 章, 大谷 淳

    映像情報メディア学会誌 : 映像情報メディア   51 ( 9 )  1997.09

    CiNii

  • バ-チャル歌舞伎における実時間表情検出

    海老原 一之, 楜沢 順, 大谷 淳

    画像ラボ   8 ( 9 ) 1 - 5  1997.09

    CiNii

  • 多数カメラによる実時間手振り推定装置を用いた対話的シーン生成

    内海章, くるみ沢順, 大塚尚宏, 大谷淳

    日本バーチャルリアリティ学会大会論文集   2nd   226 - 227  1997.09

    J-GLOBAL

  • A Study of Novel View Generation Using 2D Affine Space

    Kuntal Sengupta, Sakaguchi Tatsumi, Ohya Jun

    Proceedings of the Society Conference of IEICE   1997   278 - 278  1997.08

    CiNii

  • A Study of Real-time Estimation of Human Body Postures

    YAMADA Masanori, EBIHARA Kazuyuki, OHYA Jun

    Proceedings of the Society Conference of IEICE   1997   197 - 197  1997.08

    CiNii

  • Hand shape recognition using best-view selection

    UTSUMI Akira, OHYA Jun

    Proceedings of the Society Conference of IEICE   1997   198 - 198  1997.08

     View Summary

    筆者らは、非装着型のヒューマンインタフェースへの応用を目指して、画像処理による手振り認識の研究を進めており、多数カメラを利用することで安定な手の位置・姿勢推定が行なえることを示した。推定された手の位置・姿勢は、手形状認識の簡素化および安定化のための視点選択に利用できる。今回、推定姿勢に基づいて正面像を選択することで、フーリエ記述子による形状認識を安定化できることを確認したので報告する。

    CiNii

  • A Study of Novel View Generation from Stereo Images Using 2D Affine Space

    SENGUPTA Kuntal, SAKAGUCHI Tatsumi, OHYA Jun

    IPSJ SIG Notes. CVIM   1997 ( 70 ) 51 - 58  1997.07

     View Summary

    We present an unified theoretical framework for novel view synthesis and other applications. To start with, we have a set of real images from weakly calibrated cameras, for which we compute the dense point match correspondences. For applications like novel view synthesis, on may first solve the 3D scene reconstruction problem, followed by a view rendering process. However, errors in 3D scene reconstruction usually gets reflected in the quality of the new scene generated, so we seek a more direct method. In this report, we use the knowledge of dense point matches and their affine coordinate values to estimate the corresponding affine coordinate values in the new scene. Our technique of reprojection is extended for other applications like merging real and synthetic worlds.

    CiNii

  • 9)バーチャル歌舞伎システム(〔情報ディスプレイ研究会 ヒューマンインフォメーション研究会 マルチメディア情報処理研究会 ネットワーク映像メディア研究会 画像情報システム研究会〕合同)

    海老原 一之, 楜沢 順, 岩澤 昭一郎, 大谷 淳

    映像情報メディア学会誌 : 映像情報メディア   51 ( 6 ) 818 - 818  1997.06

    CiNii

  • Multiple-Camera Based Estimation of Human Position and Orientation

    MORI Hiroki, UTSUMI Akira, OHYA Jun, YACHIDA Masahiko

    Technical report of IEICE. Multimedia and virtual environment   21 ( 33(HIR97 39-59/NIM97 34-54) ) 21 - 26  1997.06

     View Summary

    We describe a system to estimate position and orientation of a human body. In such a system, deformation and occulusion of the human body often cause difficulties in the estimation. Our system employed multiple-camera based method to avoid these difficulty. Skeleton images detected from multiple cameras are utilized for stable estimation of both position and orientation. Furthermore, according to the estimation result, the best camera for the body side detection is selected. We confirmed the stability of those through experiments.

    CiNii J-GLOBAL

  • Multiple-Camera Based Estimation of Human Position and Orientation

    MORI Hiroki, UTSUMI Akira, OHYA Jun, YACHIDA Masahiko

    ITE Technical Report   12 ( 2 ) 139 - 144  1997.05

     View Summary

    We describe a system to estimate position and orientation of a human body. In such a system, deformation and occulusion of the human body often cause difficulties in the estimation. Our system employed multiple-camera based method to avoid these difficulty. Skeleton images detected from multiple cameras are utilized for stable estimation of both posltion and orientation. Furthermore, according to the estimation result, the best camera for the body side detection is selected. We confirmed the stability of those through experiments.

    CiNii J-GLOBAL

  • A Study of Recognizing Facial Expressions Generated by Unspecified Persons Based on Hidden Markov Models with Continuous Output Probabilities

    OTSUKA Takahiro, OHYA Jun

    IPSJ SIG Notes. CVIM   1997 ( 31 ) 39 - 46  1997.03

     View Summary

    Recognizing facial expressions is important in a communication through virtual scenes in which one's body and face could be transformed into other's ones. In this report, a method using HMM based on a mixture normal density as a output probability is described. First, features are obtained by applying Fourier transform to a vector field generated by a facial change. Then, the distribution of the features is approximated by mixture normal density. Finally, HMM is trained for each expression to be used for recognition. The recognition experiments using four subjects showed good recognition rate for expressions with fewer interpersonal variations.

    CiNii

  • A Study of Transformation of Facial Expressions Based on Expression Recognition from Temporal Image Sequences

    OTSUKA Takahiro, OHYA Jun

    Technical report of IEICE. PRMU   96 ( 599 ) 75 - 80  1997.03

     View Summary

    A method is proposed by which a facial expression on one's face is transformed to that on other person's face. In this method, the probabilities of muscle states (static, contracting, ending) are computed for each frame of an image sequence based on Hidden Markov models (HMM) trained for each expression. Then, the 3D wire-frame model is modified in proportion to the probabilities of the contraction state. We obtained good results from the experiment in which the six principal facial expressions (anger, disgust, fear, happiness, sadness, and surprise) were transformed.

    CiNii

  • Study of Real-Time Human Posture Estimation from Thermal Images

    IWASAWA SHOICHIRO, EBIHARA KAZUYUKI, OHYA JUN, MORISHIMA SHIGEO

    Technical report of IEICE. HCS   96 ( 604 ) 37 - 44  1997.03

     View Summary

    This report proposes a new real-time method that estimates the posture of a human from a thermal image acquired by an infrared camera regardless of background and lighting conditions. Distance transformation is performed for the human body area extracted from the thresholded thermal image so that the center of gravity of body is calculated. After the orientation of the upper half of the body is obtained from calculating the moment of inertia, significant points such as the top of the head, the tips of the hands and foot are heuristically located. In addition, the elbow and knee positions are estimated from the siginificant points using a genetic algorithm based learning procedure. This method does not require to be put any type of device on a body and it can be applied for arbitrary person.

    CiNii

  • Real-Time Estimation of Human Body Posture from Thermal Images

    Iwasawa Shoichiro, Ebihara Kazuyuki, Ohya Jun, Morishima Shigeo

    Proceedings of the IEICE General Conference   1997 ( 2 ) 365 - 365  1997.03

     View Summary

    筆者らは既に、熱画像を用いた非接触な人物の姿勢推定手法を提案している。しかし、筆者らの従来法では、上半身の左右への大幅な傾きや、足の広範な動作に対応できず、また単眼のため3次元情報が得られないなどの課題があった。本報告ではより多くの姿勢に対応できるよう、従来の単眼用アルゴリズムを改良する。即ち単眼赤外線カメラから入力される熱画像から獲得される人物領域と、その輪郭の情報に基づいてヒューリスティックに頭頂・手先・足先の各位置を実時間で検出する手法、および遺伝的アルゴリズム (GA) を利用して肘および膝の位置を推定する手法を提案する。さらに、ステレオ視による3次元位置の獲得について検討する。

    CiNii

  • 時系列画像から推定される画素値分布に基づく手領域抽出

    内海章, 大谷淳

    電子情報通信学会大会講演論文集   1997 ( Sogo Pt 7 ) 361  1997.03

    J-GLOBAL

  • A Study of Adaptive Background Image Selection Scheme According to Viewpoint

    Yamaguchi Tetsuya, Ebihara Kazuyuki, Ohya Jun, Iwata Makoto, Kishino Fumio

    Proceedings of the IEICE General Conference     386 - 386  1997

    CiNii

  • A Study of PC Based Multimodal Interactive Communication System

    EBIHARA Kazuyuki, OHYA Jun

    ITE Winter Annual Convention     103 - 103  1997

    CiNii

  • 5)熱画像からの実時間全身像の姿勢推定法の検討(ネットワーク映像メディア研究会)

    岩澤 昭一郎, 海老原 一之, 大谷 淳

    テレビジョン学会誌   50 ( 12 ) 1987 - 1987  1996.12

    CiNii

  • 10)造形学的視点からの表情再現([マルチメディア情報処理研究会 ネットワーク映像メディア研究会]合同)

    海老原 一之, 楜沢 順, 大谷 淳, 岸野 文郎

    テレビジョン学会誌   50 ( 11 ) 1817 - 1817  1996.11

    CiNii

  • 10)造形学的視点からの表情再現([マルチメディア情報処理研究会 ネットワーク映像メディア研究会]合同)

    海老原 一之, 楜沢 順, 大谷 淳, 岸野 文郎

    テレビジョン学会誌   50 ( 9 ) 1419 - 1419  1996.09

    CiNii

  • Toward Integration of Pattern Recognition/Understanding and Art

    Nakatsu Ryohei, Ohya Jun

    Proceedings of the Society Conference of IEICE   1996   541 - 542  1996.09

    CiNii

  • 3D Real-time Mouth Animation Controlled by Two Detected Parameters Using Bilinear Interpolation

    MOUBARAKI Lhassan, OHYA Jun

    IEICE technical report. Image engineering   96 ( 198 ) 39 - 46  1996.07

     View Summary

    Our real-time animation algorithm enhances the realism of computer-generated 3d mouth and lip motion using a complete parametric model of the mouth based only on the width and height (w,h) between the lips. The animation algorithm which is totally controlled by (w,h) performs a filtering and a calibration of the 2D input data (w,h), and computes the openingangle of the mouth and the z-component (for protrusion effects) of the motion using a bilinear interpolation. The contours of the lips are created using splines. Finally, a real-time 3D animation sequence controlled by the detected (w,h) is obtained using a parametric mouth motion.

    CiNii

  • 4)周波数領域変換を用いた実時間表情検出(ネットワーク映像メディア研究会)

    海老原 一之, 大谷 淳, 岸野 文郎

    テレビジョン学会誌   50 ( 7 ) 997 - 998  1996.07

    CiNii

  • A study of Real-time Facial Expression Reproduction Based On Artistic anatomy

    EBIHARA Kazuyuki, KURUMISAWA Jun, OHYA Jun, KISHINO Fumio

    ITEJ Technical Report   20 ( 41 ) 55 - 59  1996.07

     View Summary

    In visual communication systems in which 3D images of humans and scenes need to be reproduced so that the participants can feel as if they share one common space, fidelity of reproduction of facial expressions is very important. In this paper, we propose a new expression reproduction method based on artistic anatomy. Artists create reference facial expression 3D models and generate other 3D expressions by mixing the reference expressions. Then before telecommunications, the parameters representing the mixtures are related to the facial expression parameters to be obtained by the expression detection system so that detected expressions are reproduced by using the relationship. Promising experimental results are show.

    CiNii

  • Real - time Detection and Reproduction of Facial Expressions in Vitual Space Teleconferencing

    OHYA Jun, KISHINO Fumio

    IPSJ SIG Notes. CVIM   1996 ( 47 ) 47 - 54  1996.05

     View Summary

    Real-time facial expression detection and reproduction systems developed by the authors are introduced. In the detection method, frequency domain transform is utilized to detect deformations of facial components in face images. Facial expression are reproduced in 3D face models by deforming the models according to the expression detection data transformed based on 3D measurement data for the faces on different facial expressions. The perfomance of the detection and reproduction are shown, and remaining issues are described.

    CiNii

  • A Study of 3D Synthesis of Human Body Movements for Virtual Space Teleconferencing

    NARIYAMA Keiichi, OHYA Jun, KISHINO Fumio

    Technical report of IEICE. Multimedia and virtual environment   95 ( 553 ) 57 - 64  1996.03

     View Summary

    This paper describes realistc 3D reproduction of human body movements for virtual space teleconferencing. This paper deals with body motions above the waist and transitional movements between sitting standing and bowing. The skeletal structure of a 3D human model consists of the links connected by joints with 29 DOF. By using constraints, these joint angles are determined from the data from 4 magnetic sensors attached to the head, chest and hands. Free form deformations (FFD), which are applied to the spine, animate natural bending of torso model at a speed of 8 frames/sec..

    CiNii

  • Real-Time Facial Expression Detection Based on Frequency Domain Transform

    EBIHARA Kazuyuki, OHYA Jun, KISHINO Fumio

    ITEJ Technical Report   20 ( 15 ) 17 - 22  1996.03

     View Summary

    A new method for real-time detection of facial expressions from time-sequential images is proposed. The proposed method does not need tape marks that are pasted to the face for detecting expressions in real-time in the current implementation for Virtual Space Teleconferencing. In the proposed method, four windows are applied to four areas in the face image : a left and right eyes, mouth and forehead. Each window is divided into blocks that consist of 8 by 8 pixels. Discrete Cosine Transform (DCT) is applied to each block, and the feature vector of each window is obtained by taking the summations of the DCT energies in the horizontal, vertical and diagonal directions. To convert the DCT features to virtual tape mark movements, we represent the displacement of a virtual tape mark by a polynomial of the DCT features for the three directions. We apply a Genetic Algorithm to train facial expression image sequences to find the optimal set of coefficients that minimizes the difference between the real and converted displacements of the virtual tape marks. Exprimental results shows the effectiveness of the proposed method.

    CiNii

  • Study of Real-Time Human Posture Estimation from Thermal Images

    IWASAWA SHOICHIRO, EBIHARA KAZUYUKI, OHTANI JUN

    Technical report of IEICE. OFC   96   25 - 30  1996

     View Summary

    This report proposes a new real-time method that estimates the posture of a human from a thermal image acquired by an infrared camera regardless of background and lighting conditions. Distance transformation is performed for the human body area extracted from the thresholded thermal image so that the center of gravity of body is calculated. After the orientation of the upper half of the body is obtained from calculating the moment of inertia, significant points such as the top of the head, the tips of the hands and foot are heuristically located. In addition, the elbow and knee positions are estimated from the significant points using a genetic algorithm based learning procedure. It turns out the proposed method works effectively in the Virtual Kabuki System.

    CiNii

  • Real Time 3-D Facial Image Reconstruction Based On 3D Measurement

    EBIHARA Kazuyuki, OHYA Jun, SUZUKI Noriko, KISHINO Fumio

    Proceedings of the Society Conference of IEICE   9 ( 2 ) 527 - 536  1996

    CiNii

  • A Study of Real Time Human Motion Capture by a Thermo Graph

    Ebihara Kazuyuki, Kurumisawa Jun, Ohya Jun

    Proceedings of the Society Conference of IEICE     212 - 212  1996

    CiNii

  • A Study of Texture Mapping Method for Reconstruct 3D Objects based on Planary Symmetry Using Active Camera

    Yoshida Kazunori, Ohya Jun, Kishino Fumio

    IEICE technical report. Pattern recognition and understanding   95 ( 445 ) 25 - 30  1995.12

     View Summary

    In this paper, we propose a new color inferring method for reconstructed 3D objects based on planary symmetry using active camera. In the proposed method, the lighting condition is measured using color samples before inferring color of observed objects. From the observed RGB data of the color samples, RGB data of the object observed by the CCD camera, and the surface normal to the point on the surface of the object, the original color data of the plane is inferred.

    CiNii

  • 3-D Facial Image Processing in the Virtual Space Teleconferencing System

    EBIHARA Kazuyuki, YAMADA Masanori, OHYA Jun, KISHINO Fumio

    ITEJ Technical Report   19 ( 44 ) 21 - 26  1995.09

     View Summary

    In our virtual space teleconferencing system, in order to detect facial expressions and to reconstruct 3-D facial images, 3-D markers are attached to the face and traced visually. For the reconstruction of facial expressions, the tracing results are used to control the nodes of the wire frame model. In this paper, we explain the marker tracing system which consists of a CCD camera, a marker tracing device, and a personal computer. Markers are traced by this system at the sending site. Also we explain the facial image reconstructing system at the receiving site.

    CiNii

  • A study of full-body movement synthesis using data from a small number of the sensors

    NARIYAMA Keiichi, SINGH Karansher, OHYA Jun, KISHINO Fumio

    Proceedings of the Society Conference of IEICE   1995   183 - 183  1995.09

    CiNii

  • 1)テクスチャーモーフィングを用いた臨場感通信会議のための布の実時間アニメーション効果([放送方式研究会画像処理・コンピュータビジョン研究会映像表現研究会画像通信システム研究会画像応用研究会]合同)

    シン カランシャー, 大谷 淳, 岸野 文郎

    テレビジョン学会誌   49 ( 5 ) 704 - 704  1995.05

    CiNii

  • 3-D Facial Image Reconstruction Based on 3D Measurement

    SUZUKI Noriko, OHYA Jun, KISHINO Fumio

    Proceedings of the IEICE General Conference   1995   252 - 252  1995.03

    CiNii

  • Facial Expression Recognition from Image Sequence : Using Wavelet Transformation and HMM

    SAKAGUCHI Tatsumi, OHYA Jun, KISHINO Fumio

    Proceedings of the IEICE General Conference   1995   258 - 258  1995.03

    CiNii

  • A Study of 3D Facial Image Reconstruction According to Differential Importance of 3D Measurements.

    鈴木紀子, 大谷淳, 岸野文郎

    電子情報通信学会技術研究報告   94 ( 486(HC94 82-86) )  1995

    J-GLOBAL

  • A method for detecting and tracking complex contours by using Snakes

    HATA Koichi, OHYA Jun, KISHINO Fumio

    Proceedings of the IEICE General Conference   1995 ( 2 ) 326 - 326  1995

    CiNii

  • A study of upper-body movement synthesis using data from a small number of the sensors

    NARIYAMA Keiichi, SINGH Karansher, OHYA Jun, KISHINO Fumio

    Proceedings of the IEICE General Conference   262   262 - 262  1995

    CiNii

  • Automatic Contour Extraction and Tracking by using AIC-based Snakes

    Hata Koichi, Ohya Jun, Kishino Fumio

    IEICE technical report. Pattern recognition and understanding   95   25 - 30  1995

     View Summary

    Abstract : This papef proposes a new method for automatic extraction and tracking complex contours like human contours. We assume that a complex contour consists of contour segments whose spatial coordinates and the gradient of intensity in direction normal to themselves are smOoth. In our approach every contour segment is tracked by an Active Contour Model(ACM).In conventional methods J initial positions of ACMs need to be given manudlly. In contrast, our method can estimate the initial positons of ACMs and track contours automatically, where whether contours are connected or not is judged based on AIC. Our method can find contour segments occluded so that tracking is stopped. Experimental results using a reaI image sequence show promising res ults .

    CiNii

  • Real-time Cloth Animation Effects for Virtual Space Teleconferencing Using Texture Morphing

    SINGH Karansher, OHYA Jun, KISHINO Fumio

    ITEJ Technical Report   18 ( 64 ) 1 - 8  1994.11

     View Summary

    This paper presents techniques for the synthesis and real time animation of tight fitting apparel on human figures within a virtual space teleconferencing system. Physical modeling and animation techniques for cloth, though effective, are too complex to achieve real time results. The emphasis in teleconferencing is more on real time visual realism than physical accuracy. In our approach clothes are modeled as textures mapped onto a geometric skin. Characteristic wrinkled textures for different garments are synthesized and registered on the skin in a semiautomatic and interactive fashion. Wrinkle formation is carried out by texture morphing controlled by the posture of the animated human. Enhanced realism is illustrated within a teleconferencing system by employing hardware pixel blending to morph between synthesized cloth textures.

    CiNii

  • Human Posture Detection from Trinocular Images Using Genetic Algorithm : Posture Estimation of Upper Half Bodies and Fingers

    OHYA Jun, KISHINO Fumio

      1994   465 - 466  1994.09

    CiNii

  • A-155 Deformation Rules in 3-D Facial Image Reconstruction

    SUZUKI Noriko, OHYA Jun, KISHINO Fumio

      1994   155 - 155  1994.09

    CiNii

  • A-161 Reconstruction of Facial Action Images with Texture-Blending

    Kitamura Yasuichi, Ohya Jun, Kishino Fumio

      1994   161 - 161  1994.09

    CiNii

  • Realistic Modeling and Animation of a Muscle and Skin Layer for Human Figures using Implicit Function Techniques

    SINGH Karansher, OHYA Jun, KISHINO Fumio

    ITEJ Technical Report   18 ( 41 ) 49 - 56  1994.07

     View Summary

    This paper presents a unified muscle and skin layer model for articulated figures, allowing the application of Implicit Function and Brep (Boundary Representation) based animation techniques. Human figures are represented using a layered approach. The realistic modeling and animation of a muscle and skin layer over a skeletal structure involves issues such as ensuring smooth connectivity between the animated segments of the virtual human, modeling the formation of creases, bulges and contact deformation on collisions. Solutions using Implicit functions to solve problems resulting due to the animation of the human figure are presented. The application of the above models and ideas to enhancing the realism in a virtual space teleconferencing system is then discussed.

    CiNii

  • 5)知覚的運動幾何学 : ロボットマニピュレータの視覚に基づく制御([放送方式研究会映像表現研究会画像処理・コンピュータビジョン研究会画像通信システム研究会画像応用研究会]合同)

    クッカ ピーター, 大谷 淳, 岸野 文郎

    テレビジョン学会誌   48 ( 3 ) 356 - 356  1994.03

    CiNii

  • 7)ホモトピーに基づいた三次元顔画像の表情生成の検討([視聴覚技術研究会画像処理・コンビョン研究会]合同)

    梶原 創一, 田中 弘美, 大谷 淳, 岸野 文郎

    テレビジョン学会誌   48 ( 1 ) 116 - 117  1994.01

    CiNii

  • Homotopy-Based 3D Animation of Facial Expressions

    Moubaraki Lhassan, Tanaka Hiromi, Kitamura Yasuichi, Ohya Jun, Kishino Fumio

    IEICE technical report. Image engineering   94   9 - 16  1994

     View Summary

    The Homotopy Sweep Method was proposed in 1991 for surface generation using a set of twodimensional contours with the interesting ability to control the transition from one contour to the other.The evaluation and the implementation on Silicon Graphics showed us the limitations of this approach for human face representation.However,it was interesting to take advantage of the convenient control of the deformation by developing a new formulation based on a dynamic approach.With some assumptions concerning the muscles and the human morphology,wecan control mouth continuous deformation just by specifying a few parameters: the animation results are realistic and the simple interpolation computations can be considered as a real time process.

    CiNii

  • Human Posture Detection from Multiple Images Using Genetic Algorithm : Detection of Joint Angles of Upper Half Bodies

    Ohya Jun, Kishino Fumio

    IEICE technical report. Pattern recognition and understanding   93   47 - 54  1994

     View Summary

    A new method for detecting human upper-half body postures and positions of the humans from multiple images based on Genetic Algorithm is proposed.The parameters to be detected are corresponded to the genes,and as the fitness to environment,an overlapping area rate of real human images and sythesized human images obtained by deformin the 3D human model based on the information of the genes is used.At each generation,natural selection,crossover and mutation are performed,and after some generations,the estimated parameter values are obtained from the genes of the individual having the best fitness.Experiments using synthesized human images show promising results.

    CiNii

  • Analysis and Synthesis of Facial Expression Based on Three- dimensional Measurement

    Sakaguchi Tatsumi, Morishima Shigeo, Ohya Jun, Kishino Fumio

    Technical report of IEICE. HC     61 - 68  1994

     View Summary

    We′ve been working on the human-machine interface using facial e xpression animation.But the model based facial expression synthesis method that we proposed are not satisfaction.Because this model deformation rule is constructed based on the two dimensional measurement of human face.In this paper,we propose the three dimensional measurement method for facial surface movement and decide the new deformation rule of facial model.In this method, the three dimensional information is reconstructed from the front and side view image.Furthermore we reconsider the quantify of Action Units in FACS and the interpolation of feature points from the measurement result.

    CiNii

  • Mirror Symmetry Detection from 2.5D Images

    SATO Yoshinobu, OHYA Jun

    The Transactions of the Institute of Electronics,Information and Communication Engineers.   76 ( 11 ) 2444 - 2447  1993.11

    CiNii

  • A Study of grabbing facial actions from facial images with genetic programming

    Kitamura Yasuichi, Ohya Jun, Kishino Fumio

    IEICE technical report. Pattern recognition and understanding   93   23 - 28  1993

     View Summary

    We proposed a new technique for grabbing facial actions using genetic programming and discussed about the techinique′s performan ce.We limited each gene to a certain number of facial muscle motions and reconstructed a 3D facial model from the genetic information.The value of each model was set to the sum of the difference between the pixels of 3D reconstructed facial image and the actual facial image whose configuration information was grabbed from the subject face using a TV camera.The selection was then made using that value.Gene processing was also done in a genetic pool.We obtained the actual facial muscle parameter for a certain number of generations of these selections.We got good results with both 2D and 3D facial processing.

    CiNii

  • Time-sequential range image acquisition by stereo integration of thermal images and intensity images.

    大谷淳, 大和淳司, 石井健一郎

    電子情報通信学会大会講演論文集   1992 ( Shunki Pt 7 ) 7.396  1992.03

    J-GLOBAL

  • Human-independent Action Recognition using Hidden Markov Model from Time-Sequential Images.

    大和淳司, 大谷淳, 石井健一郎

    電子情報通信学会大会講演論文集   1992 ( Shunki Pt 7 ) 7.353  1992.03

    J-GLOBAL

  • Time-sequential range image acquisition by stereo thermal images.

    大谷淳, 大和淳司, 石井健一郎

    電子情報通信学会技術研究報告   91 ( 478(PRU91 126-136) ) 25 - 32  1992.02

    J-GLOBAL

  • Recognizing Human Action in Time-Sequential Images using HMM.

    大和淳司, 大谷淳, 石井健一郎

    電子情報通信学会技術研究報告   91 ( 252(PRU91 51-61) ) 59 - 66  1991.09

    J-GLOBAL

  • Human Action Recognition using HMM.

    大和淳司, 大谷淳, 石井健一郎

    情報処理学会全国大会講演論文集   43rd ( 2 ) 2.245-2.246  1991.09

    J-GLOBAL

  • Range image acquisition by stero thermal image.

    大谷淳, 大和淳司, 石井健一郎

    情報処理学会全国大会講演論文集   43rd ( 2 ) 2.443-2.444  1991.09

    J-GLOBAL

  • 7)3-L法を用いた普通紙への階調画像記録(画像表示研究会(第61回))

    徳永 幸生, 大谷 淳, 久保田 一成

    テレビジョン学会誌   36 ( 1 ) 76 - 76  1982.01

    CiNii

▼display all

Awards

  • The Institute of Image Electronics Engineers of Japan, Excellent Paper Award

    2012.06  

  • 15th Telecommunication Advancement Foundation Award (TELECOM System Technology Prize) by The Telecommunication Advancement Foundation (TAF)

    2000  

  • The 57th Outstanding Patent Applicationss

    1998  

Research Projects

  • Analysis of behaviors of non-rigid objects in real space and their real-time reproduction in 3D virtual space

  • Research on the real-time reproduction of real non-rigid objects' behaviors in 3-dimensional virtual space

  • Research on Reproducing Behaviors of Non-rigid Objects Present in Real Space in Virtual 3D Space

  • Research on Recognizing Human Actions from the Images Acquired by Multiple Static or Dynamic Cameras

  • Study of Analyzing Dynamical Behaviors and 3D Reconstruction from Video Sequences

  • 動画像処理による非剛体の挙動解析の研究

    科学研究費助成事業(早稲田大学)  科学研究費助成事業(基盤研究(C))

▼display all

Presentations

  • ダ―モスコープで撮像した口腔画像と深層学習を用いた口腔疾患検出法に関する検討

    鈴木優太, 大谷 淳, 岡本俊宏, 貝淵信之, 坂口勝久, 吉光喜太郎, 福沢栄治

    電子情報通信学会技術報告MI研究会, vol. 122, no. 188, MI2022-58, pp. 39-44 

    Presentation date: 2022.09

  • 少数豚画像からGANにより生成された画像を用いて訓練された深層学習を用いた豚足切断箇所の推定法

    神田浩希, 大貫 崚, 大川内隆朗, 大谷 淳

    電子情報通信学会技術報告IE研究会, vol. 122, no. 30, IE2022-19, pp. 97-102 

    Presentation date: 2022.05

  • QRコード認識機能および仮想分度器を用いたARナビゲーションによる穿刺手法

    田中陽也, 森田 賢, 鈴木一史, 山本敬洋, 山崎宙士, 藤井柊平, 大谷 淳, 正宗 賢, 坂井修二

    電子情報通信学会技術報告MI研究会, vol. 122, no. 31, MI2022-10, pp. 55-60 

    Presentation date: 2022.05

  • An Actor-Critic based Reinforcement Learning Algorithm for Combinatorial Optimization and Mobile Power Trucks Routing Problem

    Zhao Wang, Yuhei Senuma, Yuusuke Nakano, Jun Ohya, Ken Nishimatsu  [Invited]

    IEICE NS Technical Meeting, vol. 122, no. 5, NS2022-1, pp. 1-6 

    Presentation date: 2022.04

  • Deep Cascade Road Extraction Network: a Multi-task Method for Road Extraction

    Yubo Wang, Zhao Wang, Yuusuke Nakano, Ken Nishimatsu, Katsuya Hasegawa, Jun Ohya

    IEICE PRMU Technical Meeting, vol. 121, no. 427, PRMU2021-60, pp. 1-6 

    Presentation date: 2022.03

  • ドローンによる多時期空撮画像を用いた土砂災害の検出に関する研究

    山口洋祐, 松井 開, 大谷 淳, 長谷川克也, 長橋 宏

    電子情報通信学会技術報告PRMU研究会, vol. 121, no. 427, PRMU2021-71 

    Presentation date: 2022.03

  • 骨格情報と深層学習を用いた手術着着用状態における動画像中人物の識別

    藤井柊平, 長橋 宏, 正宗 賢, 大谷 淳

    電子情報通信学会技術報告PRMU研究会, vol.121, no. 427, PRMU2021-80, pp.127-132 

    Presentation date: 2022.03

  • ワイヤーハーネス製造の自動化のためのRGBD画像を用いるコネクタの嵌め込み状態と姿勢の推定に関する研究

    鎌田賢知, 福沢栄治, 大谷 淳

    電子情報通信学会技術報告IE研究会, vol. 121, no. 422, IE2021-115, pp. 222-227 

    Presentation date: 2022.03

  • 人共存型自律移動ロボットにおけるステレオカメラと機械学習を用いた周辺人物の状態認識および行動予測

    林 正晃, 大谷 淳, 大和淳司, 亀﨑允啓, 斎藤恭介, 濱田太郎, 櫻井絵梨子, 菅野重樹

    電子情報通信学会技術報告IE研究会, vol. 121, no. 422, IE2021-114, pp. 216-221 

    Presentation date: 2022.03

  • 人共存型パーソナルモビリティに搭載されたカメラにより獲得される動画像からの深層学習に基づく屋外路面の通行容易性推定

    中山瑛介, 大谷 淳, 大和淳司, 亀﨑允啓, 葛西優介, 菅野重樹

    電子情報通信学会技術報告IE研究会, vol. 121, no. 422, IE2021-116, pp. 228-233 

    Presentation date: 2022.03

  • VRとハードウェアの統合によるスキー技術獲得システムの有効性の評価

    岡田 雄士, 瀬尾 燦振, 谷口 基文, 彼末 一之, 小方 博之, 大谷 淳

    画像電子学会第299回研究会講演予稿, VOL. 21-03, 21-03-04, pp.15-22 

    Presentation date: 2022.02

  • 左心房CT画像と機械学習を用いる左心耳の血栓形成予測法に関する検討 ― VGG16とGrad-CAMによる血栓形成に関する注目領域の可視化とLight-GBMによる血栓生成要因推測 -

    三瓶 聡太, 山本 匡, 岩崎 清隆, 長橋 宏, 大谷 淳

    画像電子学会第299回研究会講演予稿, VOL. 21-03, 21-03-10, pp.46-53 

    Presentation date: 2022.02

  • RGB画像からの深層学習を用いた圃場における優勢状態植物の検出方法

    井出 春佳, 青竹 峻太郎, 小方 博之, 大谷 淳, 大谷 拓也, 高西 淳夫, 舩橋 真俊

    画像電子学会第299回研究会講演予稿, VOL. 21-03, 21-03-31, pp.186-192 

    Presentation date: 2022.02

  • Self-trainingを用いる夜間森林環境画像のセグメンテーションに関する検討

    筒井 翔太, 大和 淳司, 大谷 淳, 佐島 遼

    画像電子学会第299回研究会講演予稿, VOL. 21-03, 21-03-30, pp.178-185 

    Presentation date: 2022.02

  • 月面環境における縦穴昇降ロボットの自己位置推定手法の検討 -Visual SLAMとウインチオドメトリの組み合わせー

    鈴木 泰貴, 大和 淳司, 石井 裕之, 大谷 淳, 高西 淳夫

    画像電子学会第299回研究会講演予稿, VOL:21-03, 21-03-26,pp.153-158 

    Presentation date: 2022.02

  • 5-ALAによる蛍光画像と深層学習に基づく初期口腔がん診断に関する研究 ~ GANを用いた蛍光画像の自動生成とCNNを用いた自動的な進行度分類方法の検討 ~

    藤元太郎, 福沢栄治, 舘原誠晃, 里村一人, 大谷 淳

    電子情報通信学会技術報告MI研究会, vol. 121, no. 347, MI2021-76, pp. 135-140 

    Presentation date: 2022.01

  • Pedestrian Position Estimation by Integrating AlphaPose Based Human Pose Estimation and PredRNN++ Based Spatiotemporal Series Prediction for Autonomous Driving Systems

    Zhang Zhaowei, Eiji Fukuzawa, Jun Ohya

    Presentation date: 2021.08

  • 協生農法環境におけるRGB画像からの圃場の優勢植生の深層学習を用いる検出方法に関する研究

    征矢 寛汰, 青竹 峻太郎, 小方 博之, 大谷 淳, 大谷 拓也, 高西 淳夫, 舩橋 真俊

    2021年度第49回画像電子学会年次大会講演論文集, S7-3, pp.1-4 

    Presentation date: 2021.06

  • 協生農法環境におけるRGB画像に対するSemantic Segmentationを用いた圃場の被覆状態の認識方法に関する研究

    吉崎 玲奈, 青竹 峻太郎, 小方 博之, 大谷 淳, 大谷 拓也, 高西 淳夫, 舩橋 真俊

    2021年度第49回画像電子学会年次大会講演論文集, S8-2, pp.1-4. 

    Presentation date: 2021.06

  • UAVを用いた災害時の道路画像分割に関する研究

    武藤信太, 大谷 淳

    電子情報通信学会画像工学研究会, vol. 120, no. 390, IE2020-70, pp. 97-102 

    Presentation date: 2021.03

  • 環境モニタリングロボットを用いた夜間森林環境モニタリングにおけるSegNetを用いた環境認識と3次元マップ生成に関する研究

    金子丈朗, 大和淳司, 石井裕之, 大谷 淳, 高西淳夫

    電子情報通信学会画像工学研究会, vol. 120, no. 390, IE2020-69, pp. 91-96 

    Presentation date: 2021.03

  • Virtual Reality環境における初心者向けスキー滑走学習システムに関する研究

    宮川峻一, 瀬尾燦振, 大橋拓未, 後藤悠太, 中新かれん, 小澤 悠, 大谷 淳, 彼末一之, 小方博之

    電子情報通信学会画像工学研究会, vol. 120, no. 390, IE2020-61, pp. 52-57 

    Presentation date: 2021.03

  • Switch State Detection by MSRS and YOLOv4 and Automatic Switch Operation with a Robot Arm by Reinforcement Learning in Virtual Environments

    Li Qi, Jun Ohya, Hiroyuki Ogata

    Presentation date: 2021.03

  • Accuracy of Needle Puncture Under Augmented Reality Displayed on a Head Mounted Display with Position Automatically Adjusted by a 2D Marker

    Satoru Morita, Kazufumi Suzuki, Kenji Endo, Takahiro Yamamoto, Shuji Sakai, Shuhei Fujii, Takeshi Morishita, Ryuhei Maruyama, Jun Ohya

    Presentation date: 2020.11

    Event date:
    2020.11
    -
    2020.12
  • ドローンに搭載したカメラ等により獲得される情報を用いる土砂災害の検出法に関する基礎的検討

    松井 開, 長谷川 克也, 武藤 信太, 大谷 淳

    2020年度第48回画像電子学会年次大会講演論文集, S3-4, pp.1-4. 

    Presentation date: 2020.12

  • Classifying Pedestrian Attention for Pedestrian Vehicle Interaction Based on Human Pose Analysis

    Zhao Wentao, Jun Ohya, Zhang Zelin

    Presentation date: 2020.12

  • How to Build AR/MR Phantom for Interventional Radiology Using Medical Images

    Kenji Endo, Kazufumi Suzuki, Satoru Morita, Shuhei Fujii, Ken Masamune, Jun Ohya, Takahiro Yamamoto, Takeshi Morishita, Ryuhei Maruyama, Shuji Sakai

    RSNA2020(Radiological Society of North America), 106th Scientific Assembly and Annual Meeting 

    Presentation date: 2020.11

  • Efficiency of Using Augmented Reality Technology in Training of Central Venous Access

    Kazufumi Suzuki, Kenji Endo, Satoru Morita, Shuhei Fujii, Ken Masamune, Jun Ohya, Takahiro Yamamoto, Takeshi Morishita, Ryuhei Maruyama, Shuji Sakai

    RSNA2020(Radiological Society of North America), 106th Scientific Assembly and Annual Meeting 

    Presentation date: 2020.11

  • Multi-Mask RCNN による多層化マスクを利用した遮蔽された物体の領域分割手法の検討

    平山 慎, 堀 隆之, 大谷 淳

    2020 年 電子情報通信学会総合大会, 情報・システム講演論文集2, D-12-3, p.36 

    Presentation date: 2020.03

  • Virtual Realityを用いたプルーク滑走スキーの体験のためのスキーシミュレータシステムの構築

    宮川峻一, 瀬尾燦振, 大橋拓未, 後藤悠太, 中新かれん, 小澤悠, 彼末一之, 大谷淳, 小方博之

    日本スキー学会第30回大会公演論文集、pp.28-29 

    Presentation date: 2020.03

  • 腹膜播種の光温熱治療のための熱画像の深層学習を用いた高解像度化に関する検討

    三瓶聡太, 尾原睦月, 瀬尾燦振, 長橋 宏, 正宗 賢, 大谷 淳, 守本祐司, 原田 学

    電子情報通信学会IE研究会,信学技報, vol. 119, no. 456, IE2019-118, pp. 103-108 

    Presentation date: 2020.03

  • RGB-Dカメラ画像を用いた災害対応ロボットによる自律的なドアノブ操作方法に関する検討

    林 正晃, 神田琢也, 宮川和也, 大谷 淳, 小方博之, 橋本健二, 孫 瀟, 松澤貴司, 内藤 博, 高西淳夫

    電子情報通信学会IE研究会,信学技報, vol. 119, no. 456, IE2019-109, pp. 63-67 

    Presentation date: 2020.03

  • 環境モニタリングロボットの自己位置推定法選択のためのRGBカメラ画像に対する深層学習による空認識の検討 ~ GNSSかVisual SLAMの選択 ~

    鈴木泰貴, 金子丈朗, 林 拓哉, 森本純矢, 大和淳司, 石井裕之, 大谷 淳, 高西淳夫

    電子情報通信学会IE研究会,信学技報, vol. 119, no. 456, IE2019-98, pp. 11-16 

    Presentation date: 2020.03

  • 自動運転システムのための深層学習を用いた合成画像から実シーン画像への変換および移動経路計画の生成

    張 澤霖, 大谷 淳

    電子情報通信学会PRMU研究会,信学技報, vol. 119, no. 481, PRMU2019-80, pp. 79-84 

    Presentation date: 2020.03

  • ドローンによる視覚障碍者の誘導支援における深層学習を用いた画像からの歩行位置推定に関する研究

    森下 健史, 長橋 宏, 大谷 淳

    画像電子学会研究会,予稿19-03-13、pp.57 - 61 

    Presentation date: 2020.02

  • 術場カメラ画像を用いる脳腫瘍摘出手術の手術工程認識に関する検討

    平田 啓樹, 大谷 淳, 正宗 賢

    画像電子学会研究会,予稿19-03-11、pp.45 - 50 

    Presentation date: 2020.02

  • Microsoft HoloLensを用いたIVR支援ARナビゲーションの知覚誤差に関する研究

    藤井柊平, 森田 賢, 鈴木一史, 遠藤健二, 森下健史, 長橋 宏, 正宗 賢, 大谷 淳

    電子情報通信学会MI研究会,信学技報, vol. 119, no. 399, MI2019-95, pp. 135-139 

    Presentation date: 2020.01

  • 深層学習を用いた心電図からの大動脈弁狭窄症の識別法の提案とGrad-CAMを用いた分析

    秦 絵里香, 瀬尾燦振, 中山雅文, 岩﨑清隆, 大川内隆朗, 大谷 淳

    電子情報通信学会MI研究会, 信学技報, vol. 119, no. 399, MI2019-87, pp. 97-101 

    Presentation date: 2020.01

  • 環境モニタリングロボットのためのRGBカメラ及び深度センサを用いた路面状況認識 〜 SegNet-BasicとENetの比較および高さと曲率特徴の比較 〜

    林 拓哉, 金子丈朗, 森本純矢, 大和淳司, 石井裕之, 大谷 淳, 高西淳夫

    電子情報通信学会技術報告、PRMU研究会 

    Presentation date: 2019.10

  • ガス管内探査ロボットの視覚機能と深層学習を用いた曲管部の湾曲方向検出に関する研究

    宮岡 寛太, 大谷 淳, 徐 栄, 高西 淳夫, 石井 裕之

    第18回情報科学技術フォーラム講演論文集 

    Presentation date: 2019.09

  • Estimating the Positions of Target Objects Based on Object Recognition by Processing 3D Point Cloud Data

    Qi Li, Jun Ohya, Atsuo Takanishi, Takashi Matsuzawa, Kenji Hashimo

    第18回情報科学技術フォーラム講演論文集 

    Presentation date: 2019.09

  • 災害対応ロボットにおけるRGB-Dカメラ情報を用いたスイッチの位置・姿勢の推定

    神田琢也, 大谷淳, 小方博之, 橋本健二, 内藤博, 松澤貴司, 高西淳夫

    第18回情報科学技術フォーラム講演論文集 

    Presentation date: 2019.09

  • 災害対応ロボットWAREC-1による自律的な階段昇段のための画像処理技術の検討

    宮川和也, 大谷淳, 小方博之, 松澤貴司, 橋本健二, 高西淳夫

    第18回情報科学技術フォーラム講演論文集 

    Presentation date: 2019.09

  • フォークリフトの倉庫内の自動運転のための経路周辺の人物の動作識別に基づく危険回避法に関する検討

    吉川 晴音, 林 拓哉, 渋谷 涼太, 大和 淳司, 大谷 淳

    2019 年度 第47 回画像電子学会年次大会 

    Presentation date: 2019.06

  • バーチャルリアリティ技術を用いた第三者視点スポーツトレーニング

    宮川 峻一, 徐 燦振, 小方 博之, 大谷 淳

    2019 年度 第47 回画像電子学会年次大会 

    Presentation date: 2019.06

  • HOG特徴と線形SVMを用いたノイズが含まれる顔動画像からの口唇の検出と口内領域の認識に関する研究

    黒宮 侃熙, 石川 彦太, 平山 慎, 大谷 淳, 徐 栄, 堀 隆之

    画像電子学会第288回研究会講演予稿 

    Presentation date: 2019.03

  • 顔の動画像解析を用いた食事のペース変動に伴うストレス検出方法の検討

    矢野 翔大, 大谷 淳, 大川内 隆朗, 大和 淳司

    画像電子学会第288回研究会講演予稿 

    Presentation date: 2019.03

  • ガス管内探査ロボットの視覚情報を用いた画像変換と進行距離推定

    杉田 愛, 大谷 淳, 徐 栄, 高西 淳夫, 石井 裕之

    画像電子学会第288回研究会講演予稿 

    Presentation date: 2019.03

  • 災害現場における4肢ロボットの深層強化学習を用いた局所的経路生成手法の検討

    黒澤俊樹, 小方博之, 大川内隆朗, 松澤貴司, 大谷 淳, 橋本健二, 高西淳夫

    電子情報通信学会技術報告、PRMU研究会 

    Presentation date: 2019.02

  • QRコード型マーカによる医療用ARナビゲーション

    丸山竜平, 森田 腎, 遠藤健二, 正宗 賢, 長橋 宏, 大谷 淳

    電子情報通信学会技術報告、MI研究会 

    Presentation date: 2019.01

  • 脳腫瘍摘出手術における術者手元近辺動画像中の手の動作解析に基づく手術状況認識法の検討

    堀込貴央, 藤江宏紀, 平田啓樹, 正宗 賢, 長橋 宏, 大谷 淳

    電子情報通信学会技術報告、MI研究会 

    Presentation date: 2019.01

  • 脳外科開頭手術のためのマーカレスARを用いた3次元医用画像の頭部画像への重畳表示

    西崎雄哉, 正宗 賢, 長橋 宏, 大谷 淳

    電子情報通信学会技術報告、MI研究会 

    Presentation date: 2019.01

  • 深層学習とGrad_CAMを用いた超音波画像中の胎児部位検出と深度推定に関する研究

    石川彦太, 徐 栄, 大谷 淳, 岩田浩康

    電子情報通信学会技術報告、MI研究会 

    Presentation date: 2019.01

  • 手術具の検出および追跡による脳腫瘍摘出手術の工程認識法におけるData Augmentationのパラメータ値の検討

    藤江宏紀, 平田啓樹, 堀込貴央, 長橋 宏, 大谷 淳, 田村 学, 正宗 賢, 村垣善浩

    電子情報通信学会技術報告、MI研究会 

    Presentation date: 2019.01

  • 動画像解析による快不快推定の研究

    矢野翔大, 大谷淳, 大川内隆朗, 大和淳司

    第31回人間情報学会予稿集 

    Presentation date: 2018.12

  • 災害対応ロボットにおけるデプスカメラ情報に基づく対象物体の把持位置・方向の自動的な推定

    西川敬士, 大谷 淳, 小方博之, 高西淳夫, 松澤貴司, 橋本健二

    FIT2018 (第17回情報科学技術フォーラム) 

    Presentation date: 2018.09

  • 疾走動作の段階的改善を目的とした教師なし学習による類似動作検出法の検討

    徐 燦振, 佐羽内雅人, 小方博之, 大谷 淳

    FIT2018 (第17回情報科学技術フォーラム) 

    Presentation date: 2018.09

  • 森林環境モニタリングロボットにおける SLAM と深層学習を用いた領域推定に基づく経路計画法の検討

    森本 純矢, 林 拓哉, 藤本 裕之, 石井 裕之, 大谷 淳, 大和 淳司, 高西 淳夫

    2018年度 第46回画像電子学会年次大会 

    Presentation date: 2018.06

  • 脚型ロボットの梯子昇降のための 3 次元点群データ解析による桟の自動的な位置推定に関する検討

    宮川 和也, 西川 敬士, 大谷 淳, 橋本 健二, 高西淳夫

    2018年度 第46回画像電子学会年次大会 

    Presentation date: 2018.06

  • Hand Pose Estimation from Single Depth Images with 3D Convolutional Neural Network

    Zelin Zhang, Zhao Wang, Jun Ohya

    電子情報通信学会技術報告 

    Presentation date: 2018.01

  • 脳動脈瘤の有無と脳動脈血管形状特徴との関係の解析に関する研究

    菊地功成, 八木高伸, 徐 栄, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2017.11

  • ドローンの墜落時点推定に有効なカメラ・センサデータ処理法の基礎的検討

    秦野 勇作, 陳 怡, 大谷 淳

    2017年度 第45回画像電子学会年次大会 

    Presentation date: 2017.06

  • ガス管内探査ロボットの視覚情報を用いた進行距離推定方法に関する研究

    杉田 愛, 徐 栄, 石井 裕之, 大谷 淳, 高西 淳夫

    2017年度 第45回画像電子学会年次大会 

    Presentation date: 2017.06

  • 似顔絵自動生成法のための深層学習による顔画像中の髪領域抽出の検討

    田中 雄也, 大谷 淳, 川村 春美

    2017年度 第45回画像電子学会年次大会 

    Presentation date: 2017.06

  • 環境モニタリングロボットの視覚情報処理を用いた森林における起点への帰還法の検討

    藤本 裕之, 石井 裕之, 大谷 淳, 大和 淳司, 高西 淳夫

    2017年度 第45回画像電子学会年次大会 

    Presentation date: 2017.06

  • RGB-D動画像処理を用いた要介護者の食事状況認識に関する検討

    矢野 翔大, 藤本 篤人, 大川内 隆朗, 大谷 淳

    2017年度 第45回画像電子学会年次大会 

    Presentation date: 2017.06

  • 顔画像中の瞼の応答性を表現する制御モデルによる目の開閉度の連続的な検出法の検討

    徐 燦振, 島 聡志, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2017.03

  • 透視変換を用いたギター演奏時のネックの動画像における追跡法の検討

    本田 浩暉, 王 ?, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2017.03

  • 脳動脈瘤の有無の識別に有効な MRA 画像中の脳血管形状特徴の検討

    齋藤 健太, 菊地 功成, 八木 高伸, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2017.03

  • Support Vector Regressionに基づく3次元動画像処理による人物の動作評価法の検討

    前田尚俊, 王 しょう, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2017.03

  • 縫合手術を構成する6つの段階の動画像中の手術者の手の動きに基づき予測する方法

    坪田実和, 李 イエ, 大谷 淳

    電子情報通信学会研究技術報告 

    Presentation date: 2017.02

  • 違和感の無いアニメーション画像生成のための背景と前景のトーンカーブを用いた自動的な融合法の検討

    江守海渡, 大谷淳, 楜沢順, 川村春美

    FIT2016 (第15回情報科学技術フォーラム) 

    Presentation date: 2016.09

  • Road Vanishing Point Detection by Multi-stage Convolutional Neural Network

    Yan LI, Ye LI, Jun OHYA

    the Media Computing Conference, IIEEJ 

    Presentation date: 2016.06

  • MRA 画像を用いた脳動脈の構造復元のための血流速度依存性と血管内外の輝度値の検討

    菊地功成, 立松直倫, 八木高伸, 大谷 淳, 張澤霖

    画像電子学会研究会 

    Presentation date: 2016.03

  • 縫合手術を構成する5つの段階の動画像からの識別法の検討

    三浦朋也, 李イエ, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2016.03

  • 3次元動画像から抽出される人物の部分骨格モデルを利用する人物の行動認識法の検討

    藤本篤人, 立松直倫, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2016.03

  • SLAM とレーザープロジェクションによる飛翔体ロボットの周囲環境との幾何学的な関係の獲得法の検討

    秦野勇作, 立松直倫, 大谷 淳

    画像電子学会研究会 

    Presentation date: 2016.03

  • Computer Vision Technologies Towards the Actualization of Robotic Scrub Nurses

    Ye Li, Jun Ohya  [Invited]

    日本画像学会誌 

    Presentation date: 2016.02

  • 地域ビジネスとロボット技術

    大谷 淳, 大野邦夫

    画像電子学会 デジタルサイネージとインタラクション(DSG)研究会 第6回ワークショップ 

    Presentation date: 2015.11

  • 抽象絵画における鑑賞者の注視情報を利用した音楽生成システム;抽象絵画の特徴が鑑賞者の抽象的な発想の変化に及ぼす影響

    小楠竜也, 大谷 淳, 楜沢 順, 米村俊一

    FIT2015 (第14回情報科学技術フォーラム) 

    Presentation date: 2015.09

  • ニューラルネットワークを利用する顔画像からの似顔絵自動生成の検討 -主観評価に基づく比較検討

    郭 帥, 榎本 誠, 川村春美, 大谷 淳

    FIT2015 (第14回情報科学技術フォーラム) 

    Presentation date: 2015.09

  • 少数特徴点追跡に基づく動画像からの樹木全体の動き追跡の検討

    沈 渊超, 大谷 淳

    FIT2015 (第14回情報科学技術フォーラム) 

    Presentation date: 2015.09

  • A Method for Tracking Guitar Neck and Fingertips: Necking Tracking Robust against Occlusions Based on Geometry Analysis and Fingertips Tracking Based on Temporal Probability Map

    Zhao Wang, Jun Ohya

    FIT2015 (第14回情報科学技術フォーラム) 

    Presentation date: 2015.09

  • 超音波画像中の胎児の組織抽出のためのSuperVoxelに基づくセグメンテーション法の検討

    王 元晨, 立松直倫, 大谷 淳, 徐 栄, 石川牧子, 岩田浩康

    FIT2015 (第14回情報科学技術フォーラム) 

    Presentation date: 2015.09

  • SLAMに基づく片づけロボットによる物体の力学的操作結果の確認方法に関する検討

    高 斯, 立松直倫, 大谷 淳

    FIT2015 (第14回情報科学技術フォーラム) 

    Presentation date: 2015.09

  • 少数特徴点追跡に基づく動画像からの樹木全体の動き追跡の基礎的検討

    シン エンチョウ, 大谷 淳

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • MRI を用いた内頚動脈再構築に関する検討-MRA 画像と血流速度分布のシミュレーションの比較・考察-

    黄 キエン, 八木 高信, 立松 直倫, 大谷 淳, 菊地 功成

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • ニューラルネットワークを用いた顔画像からの似顔絵自動生成 -輪郭と目の描画方法の改良についての検討

    郭 帥, 榎本 誠, 川村 春美, 大谷 淳

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • 絵画の”美しさ”についての基礎的な検討と考察 - 葛飾北斎「富嶽三十六景」への複数被験者による基本図形のあてはめ結果の分析―

    能條 由佳, 川村 春美, 楜沢 順, 大谷 淳

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • Study of Early Recognition and Progress Level Calculation of Surgeon's Action during Suture Surgery

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • SuperVoxel を用いた超音波画像中の胎児に対するセグメンテーションの基礎的検討

    王 元晨, 立松 直倫, 大谷 淳, 徐 栄, 石川 牧子, 岩田 浩康

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • SLAM による自己位置推定と地図生成を用いた片付けロボットの基礎的検討

    高 斯, 立松 直倫, 大谷 淳

    2015年度画像電子学会第43回年次大会 

    Presentation date: 2015.06

  • 人物の関節の三次元データを用いた動作の予測の検討 〜 リー群での相対位置に基づくSkeltonの特徴とMax-Margin Early Event Detectorsを用いた手法 〜

    立松直倫, 大谷 淳, Larry S. Davis

    電子情報通信学会研究技術報告 

    Presentation date: 2015.03

  • ピアノ練習の自動教師システムを目指したRGBD 動画像中の手指追跡の基礎的検討

    邱 東, 大谷 淳

    2015年 電子情報通信学会総合大会 

    Presentation date: 2015.03

  • 抽象絵画における鑑賞者の視線行動を利用した音楽生成システムの研究 −鑑賞者の発話内容の重複と発想の広がりとの関連に関する検討−

    小楠竜也, 大谷淳, 楜沢順, 米村俊一

    2015年 電子情報通信学会総合大会 

    Presentation date: 2015.03

  • Study of a Vision Based Method for Checking the Position of Each Finger of Guitar Players - Towards the Actualization of an Autonomous Guitar Chord Teaching System ?

    Zhao Wang, Jun Ohya

    2015年 電子情報通信学会総合大会 

    Presentation date: 2015.03

  • リー群での人物の関節の相対位置に基づくSkelton の特徴とMax-Margin Early Event Detectors を用いた動作の早期認識の検討

    立松直倫, 大谷 淳, Larry S.Davis

    2015年 電子情報通信学会総合大会 

    Presentation date: 2015.03

  • Fundamental Study of Early Recognition of Surgeon’s Action during Suture Surgery from Video Sequences

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    2015年 電子情報通信学会総合大会 

    Presentation date: 2015.03

  • Recognizing Hand Actions during Suture Surgeries by using sh-SIFT algorithm

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    第32回 日本ロボット学会 学術講演会 

    Presentation date: 2014.09

  • ベイジアンネットワークを用いた店舗内の顧客の購買行動認識法における画像特徴に関する検討

    セイ エン, 堀 隆之, 大谷 淳

    FIT2014(第13 回情報科学技術フォーラム) 

    Presentation date: 2014.09

  • 仮想化粧システムに関する基礎的検討 ― 顔形状の類似度に基づく化粧顔の検索法の検討―

    張 琳, 大谷 淳

    FIT2014(第13 回情報科学技術フォーラム) 

    Presentation date: 2014.09

  • 顔画像からの線画ベースの似顔絵自動生成システムにおける主観的識別に関する検討

    呉 玉珍, 榎本 誠, 大谷 淳

    FIT2014(第13 回情報科学技術フォーラム) 

    Presentation date: 2014.09

  • 抽象絵画における鑑賞者の視線行動を利用した音楽生成システムの研究 ―鑑賞者の発想を膨らませる音楽の自動生成システムに関する予備的検討―

    小楠竜也, 大谷 淳, 楜沢 順, 米村俊一

    FIT2014(第13 回情報科学技術フォーラム) 

    Presentation date: 2014.09

  • 店舗内の消費者の購買行動の動画像処理とベイジアンネットワークによる認識法の検討

    エン セイ, 堀 隆之, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2014.09

  • 顔画像からの似顔絵自動生成システムにおける似顔絵の類似度に関する検討

    呉 玉珍, 榎本 誠, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2014.08

  • 仮想化粧システムに関する基礎的検討

    張 琳, 大谷 淳

    2014年度 画像電子学会第42回年次大会 

    Presentation date: 2014.06

  • 顔画像からの線画ベースの似顔絵の自動生成システムにおける顔パーツ表現に関する検討

    榎本 誠, 呉 玉珍, 大谷 淳

    2014年度 画像電子学会第42回年次大会 

    Presentation date: 2014.06

  • Study of Detecting the Frets and Strings on the Neck of the Guitar from RGBD Images towards the Actualization of an Autonomous Guitar Teaching System

    Zhao WANG, Ye LI, Jing YAN, Jun OHYA

    2014年度 画像電子学会第42回年次大会 

    Presentation date: 2014.06

  • 論文の執筆過程に合わせた文献推薦システムの提案

    大川内 隆朗, 大谷 淳

    2014年度 画像電子学会第42回年次大会 

    Presentation date: 2014.06

  • 抽象絵画における鑑賞者の視線行動を利用した音楽生成システムの予備的検討 〜 提案システムが抽象絵画鑑賞者の発想の多寡に与える影響 〜

    小楠竜也, 大谷 淳, 楜沢 順, 米村俊一

    電子情報通信学会技術報告 

    Presentation date: 2014.03

  • ニューラルネットを利用する顔画像からの似顔絵自動生成における最適な特徴点の場所と数の検討

    方 国勇, 榎本 誠, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2014.03

  • 動的輪郭モデルとテクスチャ解析に基づく舌体検出と舌形分類法の検討

    田 明, 羅 丹, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2014.03

  • 移動RGB-Dカメラ画像中からの移動物体の特徴点のクラスタリングの検討 〜 Supervoxelとマルチラベルのグラフカットを用いた手法の提案 〜

    立松直倫, 大谷 淳, Larry S. Davis

    電子情報通信学会技術報告 

    Presentation date: 2014.03

  • 顔表情解析に基づく顧客の購買行動認識の基本的検討 〜 ASMによる表情特徴量の取得とベイジアンネットワークのモデルの構築手法 〜

    賈 ロイ, 大谷 淳, 堀 隆之

    電子情報通信学会技術報告 

    Presentation date: 2014.03

  • Study of Recognizing Hand Actions from Video Sequences during Suture Surgeries Based on Temporally-Sectioned SIFT and Sliding Window Based Neural Networks

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    電子情報通信学会技術報告 

    Presentation date: 2014.03

  • 複数画像特徴量を用いた読唇システム ― オプテイカルフロー特徴・形状特徴・離散コサイン変換特徴の統合の検討 −

    高橋昌平, 大谷 淳

    情報処理学会研究報告 

    Presentation date: 2014.03

  • Study of Recognizing Surgeon’s Action during Suture Surgery by Using Sectioned-SIFT

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • ニューラルネットを利用する似顔絵自動生成に関する基礎的検討 - 顔特徴点と生成された似顔絵との物理的誤差に基づく最適な特徴点数の検討 ?

    方 国勇, 榎本 誠, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • 舌診における舌体検出と舌形分類法の検討

    田 明, 羅 丹, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • 抽象絵画の鑑賞者の視線行動に基づく音楽生成システムの予備的検討 ― 音楽の有無が抽象絵画鑑賞者の発想の多寡に与える影響 ―

    小楠竜也, 大谷 淳, 楜沢 順, 米村 俊一

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • オプティカルフロー特徴と形状特徴に離散コサイン変換特徴を組み合わせる動画像からの読唇法の検討

    高橋 昌平, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • ニューラルネットワークを用いた顔画像からの線画ベースの似顔絵自動生成システムの検討 ― 鼻の描画法の検討 ―

    呉 玉珍, 榎本 誠, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • 表情解析とベイジアンネットに基づく顧客の購買行動認識に関する検討

    賈 ?維, 大谷 淳, 堀 隆之

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • 画像処理とベイジアンネットワークを用いた店舗内における顧客の全身像からの購買行動認識に関する基礎的検討

    エンセイ, 堀 隆之, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • Supervoxel とマルチラベルのグラフカットを用いた移動物体の特徴点のクラスタリングの検討

    立松 直倫, 大谷 淳, Larry Davis

    電子情報通信学会総合大会 

    Presentation date: 2013.09

  • 抽象絵画における鑑賞者の視線行動を利用した自動作曲に関する研究:視線行動パターンから和声的な音楽の生成方法に関する基礎的検討

    小楠竜也, 大谷 淳, 楜沢 順

    FIT2013 

    Presentation date: 2013.09

  • 舌診における舌体検出と特徴抽出法の検討

    田 明, 羅 丹, 大谷 淳

    FIT2013 

    Presentation date: 2013.09

  • Fundamental Study of Recognizing the Surgeon's Action during Suture Surgery from the Video Sequence

    Ye Li, Jun Ohya, Toshio Chiba, Rong Xu, Hiromasa Yamashita

    FIT2013 

    Presentation date: 2013.09

  • Fundamental Study of Particle Filter based Motorbike Tracking for a Violator Detection System

    N. Khanh Do, Jun Ohya

    FIT2013 

    Presentation date: 2013.09

  • Visual-based Human Gesture Recognition Using Multi-modal Features

    羅 丹, Jun Ohya

    FIT2013 

    Presentation date: 2013.09

  • ニューラルネットを利用する似顔絵自動生成に関する基礎的検討:ASMにより検出される顔特徴と生成された似顔絵の品質との関係の検討

    方 国勇, 榎本 誠, 大谷 淳

    FIT2013 

    Presentation date: 2013.09

  • オプティカルフローの方向特徴履歴とサポートベクトルマシンを用いた読唇の基礎的検討

    高橋昌平, 大谷 淳

    FIT2013 

    Presentation date: 2013.09

  • Improving a Bayesian Network Based Recognition of Spontaneous Facial Expressions of a Person who Watches Web News − Utilizing Image Features for Blinks ?

    Chao Xu, Jun Ohya

    FIT2013 

    Presentation date: 2013.09

  • 移動Kinect画像中からの移動物体の抽出と三次元形状のセグメンテーションに関する検討 −Probabilistic Boundary Edge MapとSupervoxelを組み合わせた手法の提案−

    立松 直倫, 大谷 淳, Larry Davis

    FIT2013 

    Presentation date: 2013.09

  • 画像処理とベイジアンネットワークを用いた店舗内における顧客の全身像からの購買行動認識の基礎的検討

    Jing Yan, 堀 隆之, 大谷 淳

    FIT2013 

    Presentation date: 2013.09

  • Study of Structure from Motion Based 3D Reconstruction and Surface Mosaic of Internal Organs from an Endoscopic Video Sequence

    Mengyao Duan, Rong Xu, Jun Ohya

    FIT2013 

    Presentation date: 2013.09

  • Study of Recognizing Spontaneous Facial Expressions of a Person who Watches Web News Based on ASM and Bayesian Network

    Chao Xu, Jun Ohya

    電子情報通信学会技術報告 

    Presentation date: 2013.09

  • Study of Structure-from-motion Based 3D Reconstruction and Surface Mosaic

    Mengyao Duan, Rong Xu, Jun Ohya

    IEICE Technical Meeting 

    Presentation date: 2013.09

  • Study of Traffic Violator Detection System for Crowded Street −Fundamental Study of Improving Template Based Motorbike Tracking Method−

    Khanh N. Do, Jun Ohya

    2013年度画像電子学会第41回年次大会予稿集 

    Presentation date: 2013.06

  • 抽象絵画を観る人の視線行動を利用した自動作曲に関する研究 −抽象絵画鑑賞時の視線の予測モデルの基礎的検討−

    小楠竜也, 大谷 淳, 楜沢 順

    2013年度画像電子学会第41回年次大会予稿集 

    Presentation date: 2013.06

  • A Robust ICP Algorithm for Automatic Registration of 3D Fetal Facial Surface

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • Temporal Modified-RANSAC とSupervoxel ベースのグラフカットを用いた移動 RBGD カメラからの移動物体領域のセグメンテーションの検討

    立松直倫, 大谷 淳, Larry S. Davis

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • サポートベクターマシンを用いたRGBD 動画像からの寝返り判定に関する基礎的検討

    峰崎 雄大, 大谷 淳

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • Robust Lip Contour Localization by Modified Active Shape Model

    Li Yan, Luo Dan, Jun Ohya

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • 店舗内における消費者の行動認識のためのベイジアンネットワークモデル構築に関する研究

    鄭 曜, 大谷 淳

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • Study of Recognizing Spontaneous Facial Behaviors Generated by Persons Watching Web News by Combining ASM and Bayesian Network

    Chao Xu, Jun Ohya

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • Face Recognition Using Sparse Representation and Moments

    Ye Li, Huazhong Shu, Jun Ohya

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • A Study on 3D Point Cloud Stitching for SFM Reconstruction

    Mengyao Duan, Rong Xu, Jun Ohya

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • RGB 画像からの3次元復元結果とDepth 画像との比較による光源位置推定法に関する基礎的検討

    林田智樹, 高橋昌平, 奥屋武志, 大谷 淳, 川村春美

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • ポリゴンモデルにおける法線ベクトルを用いた曲率の高速計算

    奥屋 武志, 藪野 健, 大谷 淳, 高橋 信之

    2013 年 電子情報通信学会総合大会 

    Presentation date: 2013.03

  • 映像合成のためのRGB画像からの3次元復元結果とDepth画像との比較による光源方向推定法に関する研究

    林田智樹, 高橋昌平, 奥屋武志, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2013.03

  • サポートベクターマシンを用いたDepth動画像からの寝返り判定に関する基礎的検討

    峰崎雄大, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2013.03

  • ベイジアンネットワークを用いた店舗内における消費者の行動認識に関する研究

    鄭 曜, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2013.03

  • 複数の視点固定型パン・チルト・ズームカメラを用いた不審者検出法に関する研究

    溝口慶範, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2013.03

  • Study of Robust Facial Feature Localization by Modified Active Shape Model

    Li Yan, Luo Dan, Jun Ohya

    IEICE Technical Meeting, Pattern Recognition and Machine Understanding 

    Presentation date: 2013.03

  • 映画予告編が映画鑑賞行動に及ぼす影響の物語論に基づく分析

    伊勢田 世山, 高橋 昌平, 奥屋 武志, 大谷 淳

    画像電子学会第264回研究会講演予稿 

    Presentation date: 2013.03

  • 高彩度および低彩度 gamut を用いた灰色仮説に基づく照明光色推定の一提案

    川村春美, 米村俊一, 大谷淳, 小島明

    情報処理学会, 研究報告オーディオビジュアル複合情報処理(AVM) 

    Presentation date: 2013.02

  • A Study of Automatic 3D Fetal Face Detection By Locating Facial Features From 3D Ultrasound Images for Navigating FETO Surgeries

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    IEICE Technical Meeting, Biomedical Engineering and Medical Image Processing 

    Presentation date: 2012.10

  • 移動Kinect 画像中からの移動物体の抽出と三次元形状のセグメンテーションに関する検討

    立松直倫, 大谷淳, Larry Davis

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • A Study of Extracting 3D Facial Feature from Kinect's Image by Integrating ASM and Depth Map

    Li Yan, Luo Dan, Jun Ohya

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • Study of Extracting Multi-modal features for Recognizing Sign Language Vocabularies that Combine Facial Expressions and Hand Gestures

    Luo Dan, Jun Ohya

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • ベイジアンネットワークを用いた店舗内における消費者の行動認識に関する研究

    鄭 曜, 大谷 淳

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • e-learning における学習者の行動履歴を利用した講義改善支援システムの開発

    大川内隆朗, 大谷淳, 米村俊一, 徳永幸生

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • A Registration Between 3D Ultrasound Images and 3D Fetal Model for Locating a Fetal Mouth in a Fetal Surgical Navigation System

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • 防犯用カメラを用いた不審者検出法の基礎的検討

    溝口 慶範, 大谷 淳

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • Fundamental Study of Reconstructing Botanical Tree's 3D Models from Kinect's Images and Dynamical Behaviors of Non-leafy Trees

    Khanh N. Do, Jun Ohya

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • Incorporating Zoom Motion Estimation in Video Compression: A More General Approach for Motion Compensated Predictive Coding

    Morsalin Uz Zoha, Jun Ohya

    FIT2012(第11回情報科学技術フォーラム) 

    Presentation date: 2012.09

  • より効率的な日本式アニメ制作ソフトとその開発に関する研究

    奥屋武志, 大谷 淳, 高橋 信之, 藪野 健

    2012年度 画像電子学会第40回年次大会予稿集 

    Presentation date: 2012.06

  • RGB・距離画像からの照明光源方向の推定に関する検討

    徐 栄, 林田智樹, 大谷 淳, 川村春美

    2012年度 画像電子学会第40回年次大会予稿集 

    Presentation date: 2012.06

  • 低彩度および高彩度色領域を用いた灰色仮説に基づく照明光色推定法の一提案

    川村春美, 米村 俊一, 大谷 淳, 小島 明

    2012年度 画像電子学会第40回年次大会予稿集 

    Presentation date: 2012.06

  • e-learning における主観的難易度の補足方法に対する基礎的検討

    大川内隆朗, 大谷 淳, 米村俊一, 徳永幸生

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • A Study of Rapid Prototyping of Facial Makeup Styles

    Jieun Cho, Minje Park, Sang Il Park, Jun Ohya

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • A Study on 3D Fetal Model Construction with Oral Cavity and Airways

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • FETO手術の自動化のために胎児の口の検出法の基礎的検討

    謝 天亮, 徐 栄, 大谷 淳, 張 博, 佐藤 嘉伸, 藤江 正克

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • 抽象絵画における鑑賞者の視線行動解析と音楽作曲 への応用に関する提案

    小楠 竜也, 大谷 淳, 楜沢 順

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • Human Gesture Recognition Using Multimodal Features

    Luo Dan, Gao Hua, Hazim Kemal Ekenel, Ohya Jun

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • Real Length Measurement by Kinect

    Mengyao Duan, Yingdi Xie, Jun Ohya

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • Study of the Particle Filter Based Method for Tracking the Dynamical Behaviors of Botanical Tree Leaves in Video Sequence

    Khanh N. Do, Jun Ohya

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • ニューラルネットを用いた似顔絵自動生成に関する基礎的検討

    高野 晃, 大谷 淳, 榎本 誠

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • 分析的記憶表象領域における顔記憶の階層的構造 - デフォルメ似顔絵自動生成システムへの応用 -

    榎本 誠, 大谷 淳

    2012年 電子情報通信学会総合大会 

    Presentation date: 2012.03

  • WEB 上で動作する“●”プロンプタを用いたコミュニケーションシステム -物語をベースにした複数ユーザ間コミュニケ-ションの特性分析-

    小駒ちなみ, 米村俊一, 徳永幸生, 杉山 精, 大谷 淳

    情報処理学会第74回全国大会 

    Presentation date: 2012.03

  • Analyzing the "●" Communication System's Effect on Users' Self-Disclosures during Tele-Communication Session

    Li Jen Chen, Jun Ohya, Shunichi Yonemura, Yukio Tokunaga

    IEICE Technical Report 

    Presentation date: 2012.03

  • A Cone Based Electromagnetic 3D Ultrasound Calibration

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    IEICE Technical Report 

    Presentation date: 2012.03

  • 胎児手術支援システムのための内視鏡画像処理による胎児の口の検出法の基礎的検討

    謝 天亮, 徐 栄, 大谷 淳, 張 博, 佐藤 嘉伸, 藤江 正克

    電子情報通信学会技術研究報告 

    Presentation date: 2012.03

  • Object Tracking Using Kinect and its Application in Tracking Dynamical Behaviors of Botanical Tree Leaves

    Khanh N. Do, Jun Ohya

    電子情報通信学会技術報告 

    Presentation date: 2012.03

  • 化粧サンプルを利用した多様な仮想メイクアップ生成法に関する考察

    智恩, 朴濟, 朴相日, 大谷 淳

    電子情報通信学会技術研究報告 

    Presentation date: 2012.03

  • 分析的記憶表象領域における顔記憶の階層的構造とデフォルメ似顔絵自動生成システムへの応用

    榎本 誠, 高野 晃, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2012.03

  • ニューラルネットを用いた似顔絵自動生成に関する基礎的検討 -ニューラルネットの構成法と物理心理的類似性の検討-

    高野 晃, 榎本 誠, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2012.03

  • 抽象絵画における鑑賞者の視線行動を利用した音楽 作曲の検討

    小楠 竜也, 大谷 淳, 楜沢 順

    情報処理学会研究報告 

    Presentation date: 2012.02

  • オンデマンド型e-learningにおける履修者の学習行動を利用した主観的難易度の把握方法の基礎的検討

    大川内隆朗, 大谷 淳, 米村俊一, 徳永幸生

    電子情報通信学会技術報告 

    Presentation date: 2011.12

  • Improving Iterative Randomized Hough Transform for Automatic Detection of Fetal Head from Ultrasound Images

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    IEICE Technical Report 

    Presentation date: 2011.11

  • A Study of Automated Fetal Head Detection by Pre-processing Based on Ultrasound Image Gray Feature and Iterative Randomized Hough Transform

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    FIT2011(第10回情報科学技術フォーラム) 

    Presentation date: 2011.09

  • 抽象絵画における鑑賞者の視線行動解析と自動音楽作曲への応用に関する基礎的検討

    小楠竜也, 大谷淳, 楜沢順

    FIT2011(第10回情報科学技術フォーラム) 

    Presentation date: 2011.09

  • 球面領域判定を用いた一枚の画像からの光源位置推定に関する検討

    立松 直倫, 徐 栄, 大谷 淳, 川村 春美, 米村 俊一

    FIT2011(第10回情報科学技術フォーラム) 

    Presentation date: 2011.09

  • 移動ステレオカメラ画像中からの移動物体抽出と三次元データのセグメンテーションの検討 -SIFT特徴点を元にした前景・背景のエネルギー関数を組み込んだ手法の提案-

    立松 直倫, 大谷 淳

    FIT2011(第10回情報科学技術フォーラム) 

    Presentation date: 2011.09

  • Studies of the Kalman Filter and Particle filter Method for Tracking the Moving Botanical Tree Leaves in Video Sequence

    Khanh N. Do, Jun Ohya

    FIT2011(第10回情報科学技術フォーラム) 

    Presentation date: 2011.09

  • 灰色仮説に基づく照明光色推定法の提案

    川村春美, 米村俊一, 大谷 淳, 松浦宣彦

    2011年度画像電子学会第39回年次大会 

    Presentation date: 2011.06

  • サイバーシアターにおける観客が実人間あるいはアバターである場合の観客効果の相違に関する検討

    張 勇, 大谷 淳, 米村 俊一, 徳永 幸生

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • A Study of Segmenting Ultrasound Image Based on Bilateral Filtering and Fuzzy Clustering

    Rong Xu, Jun Ohya, Bo Zhang, Yoshinobu Sato, Masakatsu G. Fujie

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • A Study on Edge Feature and its Descriptor

    Yingdi Xie, Jun Ohya

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • 視覚障害者の歩行支援のための動的動画像からのパーティクルフィルタを用いた白線追跡法の検討

    高橋昌平, 大谷 淳

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • 3次元実写画像の彩度を考慮した立体感強調法に関する検討

    古 鎮宇, 大谷 淳, 阿部 信明, 河合 隆史

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • 人物同士のコミュニケーションを触発する“●”メディアにおける自己開示レベルの検討

    Chen Li Jen, 大谷 淳, 米村俊一, 徳永幸生

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • テンソル分解に基づく任意の視点からの人物の動作認識手法の検討

    堀 隆之, 大谷 淳, 楜沢 順

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • Study of Estimating Face Direction and Forefinger Motion for Manipulating a Virtual Object in a Large Screen - Combining Detection of Face Direction and Particle Filter Based Hand and Finger Tracking -

    呉 明翰, 大谷淳

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • Temporal Modified-RANSAC とグラフカットを用いた移動ステレオカメラ画像中からの移動物体抽出と三次元データのセグメンテーションの検討

    立松 直倫, 大谷 淳

    2011年 電子情報通信学会総合大会 

    Presentation date: 2011.03

  • 顔の輪郭画像を用いた会話システムにおける社会的スキルと注視行動に関する考察

    花田 研, 米村俊一, 徳永幸生, 杉山 精, 大谷 淳

    情報処理学会第73回全国大会 

    Presentation date: 2011.03

  • VFT -- A Tree-Based Algorithm that is Robust to Partial-Occlusions and Rotation -

    John Delgaizo, Jun Ohya

    電子情報通信学会技術報告 

    Presentation date: 2011.03

  • 3次元実写画像の立体感強調法に関する検討 〜 彩度と明度を考慮した手法の組み合わせの比較検討 〜

    古 鎮宇, 大谷 淳, 阿部信明, 河合隆史

    電子情報通信学会技術報告 

    Presentation date: 2011.03

  • Study on Manipulating a Virtual Object in a Large Screen Close to the User by Visual Analysis of Small-Scale Movements of Forefinger, Hand and Face

    Ming-Han Wu, Jun Ohya

    電子情報通信学会技術報告 

    Presentation date: 2011.03

  • 移動ステレオカメラ画像中からの移動物体抽出と三次元データのセグメンテーションの検討 〜 Temporal Modified-RANSACと色・構造・事前確率を用いたグラフカットを組み合わせた手法の提案 〜

    立松直倫, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2011.03

  • サイバーシアターにおける観客効果に関する検討 〜 観客が実人間・アバターの場合の特性の違いに関する検討 〜

    張 勇, 大谷 淳, 米村俊一, 徳永幸生

    電子情報通信学会技術報告 

    Presentation date: 2011.03

  • パーティクルフィルタによる白線追跡のためのパラメータの検討 〜 視覚障害者の歩行支援のための人物装着型カメラにより獲得される動画像からの白線認識法の検討 〜

    高橋昌平, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2011.02

  • テンソル分解に基づく多視点映像を用いた人物の動作認識手法の検討

    堀 隆之, 大谷 淳, 楜沢 順

    電子情報通信学会技術報告 

    Presentation date: 2011.02

  • 修正ランザックによる楕円検出

    謝 英弟, 大谷 淳

    画像ラボ 

    Presentation date: 2010.12

  • 意見の対立する場面における会話特性-リアル画像と輪郭画像会話システムにおける注視行動の分析-

    花田 研, 米村俊一, 徳永幸生, 杉山 精, 大谷 淳

    2010年度ヒューマンコミュニケーショングループシンポジウム論文集 

    Presentation date: 2010.12

  • 移動ステレオカメラ画像中からの移動物体の抽出と三次元形状復元およびカメラの姿勢推定を行うTemporal Modified-RANSACに基づく手法の検討

    立松直倫, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2010.12

  • Interactive Guitar: Musical Software That Teaches Guitar Through an Interactive Game Interface and Audio DSP

    John Del Gaizo, Jun Ohya

    IPSJ, SIGMUS 

    Presentation date: 2010.11

  • サイバーシアターにおいて多数の観客アバターが存在する場合の演技者と観客との間の観客効果に関する基礎的検討

    張 勇, 大谷 淳, 米村俊一

    電子情報通信学会技術報告 

    Presentation date: 2010.10

  • Interactive Guitar: Musical Software That Teaches Guitar Through an Interactive Game Interface and Audio DSP

    John Del Gaizo, Jun Ohya

    IPSJ, SIGMUS 

    Presentation date: 2010.10

  • 移動物体を含む環境における移動ステレオカメラの姿勢推定と空間の3次元復元の検討

    立松 直倫, 謝 英弟, 大谷 淳

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • Study of Human Gesture Recognition by Integrating Face and Hand Motion Features

    Luo Dan, Haz?m Kemal Ekenel, Jun Ohya

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • テンソル分解に基づく多視点映像を用いた人物の動作認識の一検討

    堀 隆之, 大谷 淳, 楜沢 順

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • An Improved KFCM Algorithm in the Segmentation of Brain MRIs

    Rong Xu, Jun Ohya

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • A System that Allows Human Hand's Short Range Movements to Point a Virtual Object in a Large Screen by Tracking the Hand by Computer Vision Based Approach

    Ming-Han Wu, Jun Ohya

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • テクスチャマッピング制御による立体感強調の基礎的検討

    古鎮宇, 大谷淳, 河合隆史

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • Studies on the Influences of Economic Responsibility under EPR and Producer’s Adaptive Strategy

    Yiping Zhao, Jun Ohya

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • Interactive Music: Teaching People Guitar Through Gaming Software

    John Del Gaizo, Jun Ohya

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • A System for the OMR of Ancient Musical Manuscripts

    Carolina Ramirez, Jun Ohya

    FIT2010(第9 回情報科学技術フォーラム) 

    Presentation date: 2010.09

  • Texture Analysis for Food Recognition

    Do Ngoc Khanh, Jun Ohya, Pishva Davar

    電子情報通信学会技術報告 

    Presentation date: 2010.06

  • 1チャネルの線画映像コミュニケーションにおける対話ストラテジーと対話特性

    米村俊一, 花田 研, 徳永幸生, 大谷 淳

    ヒューマンインタフェース学会研究報告集SIG-CE-01 

    Presentation date: 2010.05

  • Perlin Noise Generated Terrain Raycasting on the GPU

    Sven Forstmann, Jun Ohya

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • C.O.T.O. — A Multimedia Choreography System Inspired by Butoh-fu —

    Jennifer Pierce, Jun Ohya, John Del Gaizo, Ming-Han Wu

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • A Study of Classifying Car Types Using Front and Side Views

    Sebastien Briois, Jun Ohya

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • テンソル分解法に基づくHOG 特徴量を用いた人物の動作認識手法の検討

    堀 隆之, 大谷 淳, 楜沢 順

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • 動物体を含むパン・ティルト映像からのカメラ回転角の推定の検討

    栃折泰史, 江畑 勝, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • 色分け衣服を用いた動画像からの人物姿勢推定法の検討 — パーティクルフィルタにおける"溢れ現象"を利用した人体パーツ追跡法の提案 —

    清水雄太, 大谷淳, 岡崎潤

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • 移動カメラ画像からの手振り動きの検出と認識モデルの構築

    羅 丹, 大谷 淳

    電子情報通信学会総合大会 

    Presentation date: 2010.03

  • "●"プロンプタを用いるコミュニケーションシステムにおけるメッセージ表現の分析

    飯田真也, 米村俊一, 徳永幸生, 杉山 精, 大谷 淳

    情報処理学会第72回全国大会 

    Presentation date: 2010.03

  • 輪郭映像を用いた会話システムにおける顔注視行動の分析

    花田 研, 米村俊一, 徳永幸生, 杉山 精, 大谷 淳

    情報処理学会第72回全国大会 

    Presentation date: 2010.03

  • Policy Design and Producer’s Strategy under Extended Producer Responsibility: Economic Responsibility

    Yiping Zhao, Jun Ohya

    IPSJ SIG Technical Report 

    Presentation date: 2010.03

  • 創造的思考を用いたコミュニケーション方法 - “●”プロンプタ・システムでの実用について -

    Chen Li Jen, 米村俊一, 大谷 淳, 徳永幸生

    電子情報通信学会技術研究報告 

    Presentation date: 2010.03

  • Study of Recognizing Multiple Persons' Complicated Hand Gestures from an Active Image Sequence 〜 Increasing the Number of Persons and Hand Gesture Categories 〜

    Luo Dan, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2010.03

  • Study of Extracting Multiple Independently Moving Objects and Estimating the Egomotion from Active Stereo Sequences

    Yingdi Xie, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2010.03

  • Study of a Car Type classifying Method using an automated Frame segmentation Algorithm -- Study of a specific Representation for Convertibles using local Maxima curvature Function Features ?

    Sebastien Briois, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2010.03

  • パン・ティルト・カメラにより獲得される動画像からのカメラの回転パラメータの推定法に関する検討

    栃折泰史, 大谷 淳, 江畑 勝

    電子情報通信学会技術報告 

    Presentation date: 2010.03

  • 色分け衣服を用いた動画像からの人物姿勢推定のための人体パーツ追跡法の検討 〜 パーティクルフィルタによるオクルージョン判定法の検討 〜

    清水雄太, 大谷 淳, 岡崎 潤

    電子情報通信学会技術報告 

    Presentation date: 2010.03

  • “●”プロンプタを用いるコミュニケーションシステムにおける会話の分析 —GTAによる“●”利用プロセスの分析—

    飯田真也, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    2009年度ヒューマンコミュニケーショングループシンポジウム論文集 

    Presentation date: 2009.12

  • 輪郭映像システムを用いた説得場面における顔構成要素の役割に関する考察

    花田研, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    2009年度ヒューマンコミュニケーショングループシンポジウム論文集 

    Presentation date: 2009.12

  • カオス(現象)を用いた音楽からリアルタイムに映像を創るアート

    楜沢 順, 下西風澄, 栃折泰史, 立松直倫, 大谷 淳

    第14 回日本バーチャルリアリティ学会大会論文集 

    Presentation date: 2009.09

  • Choreography of the Butoh Object

    Jennifer Pierce, Jun Ohya

    第 14 回日本バーチャルリアリティ学会大会論文集 

    Presentation date: 2009.09

  • Ecological-value Based End-of-life Product Collecting Network Re-design and Optimization

    Yiping Zhao, Jun Ohya

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • 複数カメラを用いたMotion Signatureに基づく人物のトラッキングの検討

    堀 隆之, 大谷 淳, 楜沢 順

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • Proposed Method for OMR of Square Notation Manuscripts

    Carolina Ramirez, Jun Ohya

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • Comparison of several Shape Descriptors in order to perform Car Recognition from a moving Camera

    Sebastien Briois, Xie Yingdi, Jun Ohya

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • 修正RANSACに基づく楕円検出手法の性能評価に関する検討

    張 勇, 謝 英弟, 大谷 淳

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • 色分け衣服を用いた人物の姿勢推定法における人体パーツ追跡法の基礎的検討

    清水雄太, 大谷 淳

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • Automatic shape classification through descriptor matching

    謝 英弟, 大谷 淳

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • 回転運動するカメラにより撮影された動画象からのカメラ回転角の導出法の検討

    栃折泰史, 江畑 勝, 大谷 淳

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • 移動カメラ動画像からの手振り動作認識法の多人数動作への拡張法の検討

    羅 丹, 大谷 淳

    FIT2009(第8回情報科学技術フォーラム) 

    Presentation date: 2009.09

  • ニンフェアール第5回公演 〜息の領域〜

    カミラ・ホイテンガ, 森川栄子, 伊藤美由紀, 大村久美子, 松本祐一, 楜沢 順, 伊庭 崇, 成見 哲, 栃折泰史, 立松直倫, 大谷 淳,他

    サントリー音楽財団推薦コンサート 

    Presentation date: 2009.06

  • 画像処理を用いたサフランめしべ自動採集法の検討

    高野 晃, 小方 博之, 村松 大吾, 大谷 淳

    ロボティクス・メカトロニクス講演会2009 (ROBOMEC 2009)予稿集[No.09-4] 

    Presentation date: 2009.05

  • 顔構成要素の輪郭映像が遠隔対話に与える影響に関する考察

    花田研, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    ヒューマンインタフェース学会研究報告集 

    Presentation date: 2009.05

  • サイバーシアターにおける観客効果に関する基礎的検討

    劉 強, 大谷 淳

    2009年電子情報通信学会総合大会 

    Presentation date: 2009.03

  • テンソル分解法に基づく画像処理による個人認証における画像特徴量の比較検討

    堀 隆之, 大谷 淳, 楜沢 順

    2009年電子情報通信学会総合大会 

    Presentation date: 2009.03

  • テーブルゲーム時の競技者の顔画像からの視線検出法の検討

    唐 華斌, 大谷 淳

    2009年電子情報通信学会総合大会 

    Presentation date: 2009.03

  • A Local Method for Ellipse Detection

    Yingdi Xie, Jun Ohya

    2009年電子情報通信学会総合大会 

    Presentation date: 2009.03

  • Creative Thinking Support Communication System with Random Prompter “●” that Allows Hand-drawing

    Li Jen Chen, Shunichi Yonemura, Jun Ohya, Yukio Tokunaga

    2009年電子情報通信学会総合大会 

    Presentation date: 2009.03

  • Local Adapaptive Thresholding Study for Early Musical Mananuscripts OMR

    Carolina Ramirez, Jun Ohya

    2009年電子情報通信学会総合大会 

    Presentation date: 2009.03

  • コミュニケーションにおける会話特性に着目した顔構成要素の役割

    花田 研, 徳永幸生, 杉山精, 米村俊一, 大谷 淳

    情報処理学会第71回全国大会 

    Presentation date: 2009.03

  • “●”プロンプタを用いるコミュニケーションシステムにおける表現特性の分析

    飯田真也, 徳永幸生, 杉山精, 米村俊一, 大谷 淳

    情報処理学会第71回全国大会 

    Presentation date: 2009.03

  • 自動会話システムにおけるフォントによる感情表現の効果

    野原雄紀, 徳永幸生, 杉山精, 米村俊一, 大谷 淳

    情報処理学会第71回全国大会 

    Presentation date: 2009.03

  • Stakeholder System and Analysis of Extended Producer Responsibility Based on Game Theory

    Yiping Zhao, Jun Ohya

    情報処理学会研究報告 

    Presentation date: 2009.03

  • サイバーシアターにおける観客の表現法に関する検討

    劉 強, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2009.03

  • 2方向顔画像からのテーブルゲーム競技者の視線検出法に関する検討

    唐 華斌, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2009.03

  • Ellipse Recognition and its Application toward Medical Imaging via Edge Orientation Feature with A Modified RANSAC

    謝 英弟, 大谷 淳

    電子情報通信学会技術報告 

    Presentation date: 2009.03

  • Staff Removal for Square Musical Notation Manuscripts OCR: An Experiment on the Binarization and Staff Detection Stage

    Carolina Ramirez, Jun Ohya

    電子情報通信学会技術報告 

    Presentation date: 2009.01

  • テンソル分解に基づく人物の歩行動作を用いた個人認証

    堀 隆之, 大谷 淳, 楜沢 順

    電子情報通信学会技術報告 

    Presentation date: 2008.12

  • 画像処理による未知人物の行動認識の研究

    堀隆之, 大谷淳

    国際情報通信研究科 公開研究会 

    Presentation date: 2008.10

  • 視覚情報処理の発展と仮想コミュニケーション環境の未来

    大谷 淳

    第38回高柳記念未来技術フォーラム 

    Presentation date: 2008.09

  • The Prototype System of Prompter Communication with Handwriting and Drawing Functions

    Li Jen Chen, Sven Forstmann, Jun Ohya, Shunichi Yonemura, Yukio Tokunaga

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • Application of Synergetic Theory in Municipal Renewable Resource -- Net Values Model of End-of-life Products

    Yiping Zhao, Yingdi Xie, Jun Ohya

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • 顔画像からの顔方向推定のための鼻検出法の比較検討

    唐 華斌, 大谷 淳

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • Study of Detecting Primitive Shapes by Local Optimization Method

    Yingdi Xie, Jun Ohya

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • 複数カメラ画像からの色分け衣服装着人物の3 次元姿勢推定についての検討

    岡崎 潤, 大谷 淳

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • Study of Optimizing the Parameters in the Weighted Histogram Equalization for Stereoscopic Enhancement

    Xiao Ding, Jun Ohya, Takashi Kawai, Nobuaki Abe

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • シルエット特徴量のテンソル分解に基づく動画像から人物の動作認識手法の検討

    堀 隆之, 楜沢 順, 大谷 淳

    FIT2008(第7回情報科学技術フォーラム) 

    Presentation date: 2008.09

  • 複数カメラ画像からの色分け衣服装着人物のオクルージョンが発生する姿勢を含む姿勢推定法の検討

    岡崎潤, 大谷淳

    電子情報通信学会技術報告 

    Presentation date: 2008.09

  • Study of Stereoscopic Constrast Enhancement Based on Local Contrast and Distance Information

    Xiao Ding, Jun Ohya, Takashi Kawai, Nobuaki Abe

    電子情報通信学会技術報告 

    Presentation date: 2008.07

  • 互いに類似した動作のテンソル分解に基づく動画像からの認識法の検討

    堀 隆之, 大谷 淳, 楜沢 順

    電子情報通信学会技術報告 

    Presentation date: 2008.07

  • e-learning における講義ビデオのスキミング手法に関する提案と実装

    大川内隆朗, 大谷淳

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • スポーツ映像において高速に移動するボールを検出する方法の基礎的検討

    劉 強, 謝 英弟, 大谷 淳

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • Study of 3D Contrast Enhancement Based on Weighted Histogram Equalization

    Xiao Ding Jun Ohya, Takashi Kawai, Nobuaki Abe

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 移動カメラ画像からの3次元物体形状復元におけるカメラの移動方向評価尺度に関する検討

    成澤 祐, 大谷 淳, 小方博之

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • Detecting Ellipses from Image by Randomized Hough Transform with Edge Orientation Difference Curve Based Clustering

    Yingdi Xie Jun Ohya

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • Visualizing Large Compressed Volume Data Using a CUDA Compatible GPU

    Sven Forstmann Jun Ohya

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 移動カメラ画像からの人物追従局所座標による手振り軌跡の検出法

    羅丹, 大谷淳

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 色分け衣服装着人物の実写画像に対するMean-Shift アルゴリズムによる色領域分割の検討

    岡崎潤, 大谷淳, Huimin Guo

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 実写動画像からの人物の動作認識のために有効な特徴量の検討

    王偉卿 大谷淳

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 近赤外線画像を用いた人物の鼻の位置からの視線推定法の基礎的検討

    唐 華斌, 大谷 淳, 大川内隆朗

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 顔の特徴を分類化および言語表現可能とした似顔絵システム -デフォルメ似顔絵自動生成システム:ACGS ?

    榎本 誠, 大谷 淳, 楜沢 順

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • A Fundamental Study of “Bi-Conversation” Using “●” Communication Method

    Li Jen Chen, Shunichi Yonemura, Jun Ohya, Yukio Tokunaga

    2008年電子情報通信学会総合大会 

    Presentation date: 2008.03

  • 自動会話システムにおけるキャラクタの可変表示効果

    野原雄紀, 徳永幸生, 杉山精, 米村俊一, 大谷淳

    情報処理学会第70回全国大会 

    Presentation date: 2008.03

  • ”●“プロンプタを用いたコミュニケーションシステムの会話特性

    飯田真也, 徳永幸生, 杉山精, 米村俊一, 大谷淳

    情報処理学会第70全国大会 

    Presentation date: 2008.03

  • 顔の輪郭画像を用いた会話システムにおける視覚特性

    吉田亨, 米村俊一, 徳永幸生, 杉山精, 大谷淳

    情報処理学会第70全国大会 

    Presentation date: 2008.03

  • A Method for Creative Communication Usin Black-Circled Prompters "●" - Analysis of a Method that Allows Users' Freehand Writing -

    Li Jen Chen, Jun Ohya, Shunichi Yonemura, Yukio Tokunaga

    電子情報通信学会技術報告 

    Presentation date: 2008.03

  • Multiple Shapes Extraction with Guidance to Random Sampling

    Yingdi Xie, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2008.03

  • Visualizing run-length-encoded volume data on modern GPUs

    Sven Forstmann, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2008.03

  • モンテカルロ法とベイズ推定に基づく動的カメラからの三次元形状復元を利用したカメラの移動方向評価尺度に関する検討

    成澤 祐, 大谷 淳, 小方博之

    電子情報通信学会技術研究報告 

    Presentation date: 2008.03

  • Image Contrast Enhancement Based on Distance Information

    Xiao Ding, Jun Ohya, Takashi Kawai, Nobuaki Abe

    電子情報通信学会技術報告 

    Presentation date: 2008.03

  • Comparative Study of Methods for Recognizing Human Actions from a Real Video Sequence

    Weiqing WANG, Jun OHYA

    電子情報通信学会技術報告 

    Presentation date: 2008.03

  • 移動カメラ動画像からのCondensation Algorithmを用いた手振り認識の検討

    羅丹,大谷淳

    電子情報通信学会技術報告 

    Presentation date: 2008.03

  • 顔の特徴を分類化および言語表現可能とした似顔絵システム - A Caricatured Portrait Generate System Which Can Categorize and Describe Facial Feature -

    榎本誠 大谷淳

    日本顔学会誌 

    Presentation date: 2007.09

  • 自動会話システムにおける背景画像による話題支援

    野原雄紀, 徳永幸生, 杉山精, 米村俊一, 大谷淳

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • 動的カメラ画像からの動物体追跡のためのロボットプラットフォームの検討

    羅丹, 謝英第, 大谷淳

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • 画像処理を用いた店舗内の顧客の商品選定時における態度の認識手法の一考察

    堀 隆之, 楜沢 順, 大谷 淳

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • Study of Foreground-Background Contrast Enhancement in Color Image

    Xiao DING, Jun OHYA

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • カメラの上下動を考慮するモンテカルロ法とベイズ推定に基づく物体の三次元形状復元法の検討

    成澤 祐, 小方博之, 大谷 淳

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • 顔認識における二重符号化理論とデフォルメ似顔絵との関連性について -コンピュータによる似顔絵生成への応用-

    榎本 誠, 大谷 淳, 楜沢 順

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • Low Level Feature Detection based on Modified Ransac

    Yingdi Xie, Jun Ohya

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • VISUALIZATION OF LARGE RLE-ENCODED VOXEL VOLUMES

    Sven Forstmann, Jun Ohya

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • Tensor Decomposition Framework for Recognizing an Unknown Person’s Action from A Video Sequence Using Image Features

    Acep Irawan, Yingdi Xie, Jun Ohya

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • Validation of the Snake Pedal Model applied to Face Segmentation in Monochromatic Images

    Carolina Ramirez, Claudio Perez, Jun Ohya

    FIT2007(第6回情報科学技術フォーラム) 

    Presentation date: 2007.09

  • A Modified RANSAC Based Method for Extracting Primitive Shapes from Image

    Yingdi Xie, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2007.09

  • Study of Different Image Features Used for Recognizing Unknown Person's Activity Based on Tensor Decomposition Analysis

    Acep Irawan, Yingdi Xie, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2007.09

  • Comparative Study of Segmentation in Monochrome Facial Images: Snake Pedals and Template Matching

    Carolina Ramirez, Claudio Perez, Jun Ohya

    IEICE Technical Report 

    Presentation date: 2007.09

  • 線画映像による遠隔サポートシステム

    米村俊一, 吉田亨, 徳永幸生, 大谷淳

    日本人間工学会第48回大会 

    Presentation date: 2007.06

  • 線画によるデフォルメ映像を用いた遠隔サポートの会話特性

    米村俊一, 吉田亨, 徳永幸生, 大谷淳

    電子情報通信学会技術報告 

    Presentation date: 2007.05

  • Implementation of a Video Sequence Based System for Recognizing an Unknown Person’s Gesture by a Tensor Decomposition Based Approach

    Acep Irawan, Xie Yingdi, Jun Ohya

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • 分析的記憶表象領域における顔記憶の階層的構造とそのコンピュータシミュレーション - デフォルメ似顔絵自動生成システムへの応用 -

    榎本 誠, 大谷 淳, 楜沢 順

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • 観察法に基づく画像処理を用いた顧客の行動認識手法の一考察

    堀 隆之, 楜沢 順, 大谷 淳

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • Study of Contrast Enhancement Based on Shadow Detection

    Xiao Ding, Jun Ohya

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • Stability Improvement for Motion Segmentation Based on Active Stereo Cameras

    Yingdi Xie, Jun Ohya

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • Marker-based PC Detection Using Sample Matching

    Qi Zhang, Jun Ohya, Shunichi Yonemura

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • A Study of Communication Prompted by “●” Communication System - Approach on how prompter effects and inspires ideas and thoughts -

    Li Jen Chen, Nobuyuki Harada, Jun Ohya, Shunichi Yonemura, Yukio Tokunaga

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • Procedural Spline-Skeletons for Organic Structures and Adaptive Architecture

    Sven Forstmann, Jun Ohya

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • A Real-time 3D Human Body Tracking and Modeling System

    Huimin Guo, Jun Ohya, Ping Guo

    2007年電子情報通信学会総合大会 

    Presentation date: 2007.03

  • “●”プロンプタを用いるコミュニケーションシステム

    原田宣之, 徳永幸生, 米村俊一, 大谷 淳

    情報処理学会第69回全国大会 

    Presentation date: 2007.03

  • 自動会話システムにおける会話破綻の分析

    野原雄紀, 徳永幸生, 米村俊一, 大谷 淳

    情報処理学会第69回全国大会 

    Presentation date: 2007.03

  • 顔の輪郭映像を用いた会話システムにおける会話特性

    吉田 亨, 徳永幸生, 米村俊一, 大谷 淳

    情報処理学会第69回全国大会 

    Presentation date: 2007.03

  • 陰影領域と非陰影領域に基づく画像コントラスト強調

    丁 皛, 大谷 淳

    SCCC学会研究会報告 

    Presentation date: 2007.03

  • Skeletal Animation by Spline aligned Deformation on the GPU

    Sven Forstmann, Jun Ohya

    電子情報通信学会技術報告, 画像工学研究会 

    Presentation date: 2007.03

  • Study of Image Contrast Enhancement Based on Shadow and Non-shadow Area

    Xiao Ding, Jun Ohya

    電子情報通信学会技術報告, 画像工学研究会 

    Presentation date: 2007.03

  • A Study of Computer Mediated Communication via the “●” Prompter System - Analysis of the affects on the simulation of thought processes and the inspiration of creative ideas

    Li Jen Chen, Nobuyuki Harada, Shunichi Yonemura, Jun Ohya, Yukio Tokunaga

    電子情報通信学会技術報告, ヒューマンコミュニケーション基礎研究会 

    Presentation date: 2007.03

  • Comparative Study of Marker based Detection and Tracking of PC in the Video Sequence Acquired by the Mobile Camera for Visual Call Center Applications

    Qi Zhang, Jun Ohya, Shunichi Yonemura

    電子情報通信学会技術報告(IE) 

    Presentation date: 2007.03

  • Adaptive Optimization to Modified Ransac Based Motion Segmentation

    Yingdi Xie, Jun Ohya

    電子情報通信学会技術報告(IE) 

    Presentation date: 2007.03

  • A modified RANSAC mechanism and its application

    Yingdi Xie, Jun Ohya

    国際情報通信研究科公開研究発表会予稿集 

    Presentation date: 2006.10

  • 顔の特徴を分類化および言語表現可能とした似顔絵システム

    榎本 誠, 大谷 淳

    第11回日本顔学会大会:フォーラム顔学2006、日本顔学会誌 

    Presentation date: 2006.10

  • Independent motion areas extraction from a modified RANSAC mechanism

    Yingdi Xie, Jun Ohya

    電子情報通信学会技術報告(PRMU) 

    Presentation date: 2006.10

  • Optimization of 3D Egomotion Computation in Image Sequence

    Yingdi Xie, Jun Ohya

    FIT2006(第5回情報科学技術フォーラム) 

    Presentation date: 2006.09

  • 分析的記憶表象領域における顔記憶の階層的構造とそのコンピュータシミュレーション - デフォルメ似顔絵自動生成システムへの応用 -

    榎本 誠, 大谷 淳, 楜沢 順

    FIT2006(第5回情報科学技術フォーラム) 

    Presentation date: 2006.09

  • Study of Color Contrast Enhancement Based on Wavelet

    Ding Xiao, Jun Ohya

    FIT2006(第5回情報科学技術フォーラム) 

    Presentation date: 2006.09

  • 多関節弾性物体モデルのための計算コストと使用メモリ削減法の検討

    渡辺隆史, 大谷 淳, 楜沢 順, 徳永幸生

    FIT2006(第5回情報科学技術フォーラム) 

    Presentation date: 2006.09

  • Object Detection by Using Marker Assistance

    Qi Zhang, Jun Ohya, Shunichi Yonemura

    FIT2006(第5回情報科学技術フォーラム) 

    Presentation date: 2006.09

  • Analysis of Communication Prompted by Black “●”Communication System - Effects of Emotional Emphasis in Comparison with E-mails and Messengers -

    Li Jen Chen, Nobuyuki Harada, Jun Ohya, Yukio Tokunaga, Shunichi Yonemura

    FIT2006(第5回情報科学技術フォーラム) 

    Presentation date: 2006.09

  • Study of Real-time Marker-based PC Detection for Visual Call Center Applications

    Zhang Qi, Ohya Jun, Yonemura Shunichi

    電子情報通信学会技術報告(PRMU) 

    Presentation date: 2006.09

  • Independent motion areas extraction from a modified RANSAC mechanism

    謝 英弟, 大谷 淳

    電子情報通信学会技術報告(PRMU) 

    Presentation date: 2006.09

  • Study of Color Image Enhancement Based on Wavelet Analysis

    Ding Xiao, Jun Ohya

    電子情報通信学会技術報告(PRMU) 

    Presentation date: 2006.09

  • A Modified RANSAC mechanism: Multiple models extraction algorithm

    Yingdi Xie, Jun Ohya

    情報処理学会研究報告,Computer Vision and Image Media 

    Presentation date: 2006.09

  • 輪郭映像を用いたコミュニケーションの視覚特性

    吉田 亨, 徳永幸生, 米村俊一, 大谷 淳

    ヒューマンインタフェースシンポジウム2006 

    Presentation date: 2006.09

  • デフォルメ映像を用いた遠隔サポートのコミュニケーション

    米村俊一, 吉田 亨, 徳永幸生, 大谷 淳

    電子情報通信学会技術報告(HCS) 

    Presentation date: 2006.08

  • デフォルメ映像通信を用いた初心者ユーザの遠隔サポート

    米村俊一, 吉田 亨, 徳永幸生, 大谷 淳

    電子情報通信学会技術報告(HCS) 

    Presentation date: 2006.05

  • 顔の輪郭画像を用いた映像通信システムにおける会話特性

    吉田亨, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会第68回全国大会 

    Presentation date: 2006.03

  • 自動会話システムにおける話題切り替えの効果

    河野恭佑, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会第68回全国大会 

    Presentation date: 2006.03

  • センサーネットワークを用いる消火活動支援システムのための、画像処理による煙の挙動認識法の検討

    野飼雅弘, 大谷

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • 店舗内での顧客の行動を認識するための色情報を用いた物体の存在確認手法の基礎的検討

    堀 隆之, 大谷 淳, 楜沢 順

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • 弾性物体のように振舞うキャラクターモデルの構築法

    渡辺 隆史, 大谷 淳, 楜沢 順, 徳永 幸生

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • 移動カメラを用いるベイズ更新に基づく逐次的3次元形状復元法の検討 - 特徴点の追跡が不要な方法の提案と基礎的検討 -

    成沢 祐, 小方博之, 大谷 淳

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • 顔の特徴を分類化および言語表現可能とした似顔絵システムの提案

    榎本 誠, 大谷 淳, 楜沢 順

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • Study of Motion Segmentation from Dynamic Stereo Images Acquired by Active Stereo Cameras

    Yingdi Xie, Jun Ohya

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • Study of Designing a Prototype System for Exhibiting Artistic Works with Their Creation Processes

    Ton Srichompoo, Jun Ohya, Jun Kurumisawa, Makoto Enomoto

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • A Study of Communication Prompted by the “●” Communication System - Comparing "with Prompters" with "without Prompters" -

    Li Jen Chen, Jun Ohya, Yukio Tokunaga, Shunichi Yonemura

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • Optimise the parameters of spring models so that the Spring models' motions fit well to the real objects' motions

    Denphol Pornchanoknart Jun Ohya

    2006 年電子情報通信学会総合大会 

    Presentation date: 2006.03

  • “ANALYSIS OF HUMAN-TO-HUMAN COMMUNICATIONS PROMPTED BY THE "●" COMMUNICATION SYSTEM - Comparing "with Prompters" and "without Prompters" -

    Li Jen Chen, Jun Ohya, Yukio Tokunaga, Shunichi Yonemura

    電子情報通信学会技術報告 MVE HCGシンポジウム 

    Presentation date: 2006.03

  • Study of the Creation Processes-Based System for Exhibiting Artistic Works

    Ton SRICHOMPOO, Jun OHYA, Jun KURUMISAWA, Makoto ENOMOTO

    電子情報通信学会技術報告 MVE HCGシンポジウム 

    Presentation date: 2006.03

  • 顔の特徴を分類化および言語表現可能とした似顔絵システムの検討

    榎本 誠, 大谷 淳, 楜沢 順

    電子情報通信学会技術報告、MVE(マルチメディア・仮想環境基礎研究会) 

    Presentation date: 2006.03

  • 店舗内における顧客の行動認識方法の基礎的検討-人物の介入するフレームの検出と色情報を用いた人物と物体との接触確認-

    堀隆之, 大谷淳, 楜沢順

    電子情報通信学会、画像工学研究会 

    Presentation date: 2006.03

  • センサーネットワークを利用する消火活動支援システムの検討 - 画像処理を用いた火災状況認識法の検討 -

    野飼 雅弘, 大谷 淳

    電子情報通信学会、パターン認識・メディア理解研究会 

    Presentation date: 2006.03

  • 複数弾性物体を用いた対話操作が可能な弾性物体キャラクターモデルの構築法の提案

    渡辺隆史, 大谷 淳, 楜沢 順, 徳永幸生

    情報処理学会研究報告 

    Presentation date: 2006.02

  • Visualization of Large Caved Terrains

    Sven Forstmann, Jun Ohya

    電子情報通信学会技術研究報告 

    Presentation date: 2006.02

  • Interactive Visualization of Large ISO-Surfaces

    Sven Forstmann, Jun Ohya

    国際情報通信研究科公開研究発表会予稿集 

    Presentation date: 2005.10

  • Comparative Study of Non-rigid Objects' Animations by Spring Models and Boundary Element Method

    Denphol Pornchanoknart, Takafumi Watanabe, Jun Ohya

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • 画像情報を用いたサイバーアクターの動作制御

    Jieun Cho, 渡辺隆史, 大谷 淳

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • Study of Black "●" Prompter Communication System

    Li Jen Chen, Jun Ohya, Yukio Tokunaga, Shunichi Yonemura

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • サイバーシアターの空間移動ディレクション装置の提案

    キムヒョンムン, 大谷 淳, 小方博之

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • Study of recognizing an unknown person action from motion capture data based on tensor decomposition

    Kalanov Rovshan, Jun Ohya, Hiroyuki Ogata, Hajime Takada

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • Interactive Visualization of Large ISO-Surfaces

    Sven Forstmann, 大谷 淳

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • Study of Augmented Reality Based Method for Exhibiting Artistic Works with their Creation Processes

    Srichompoo Ton, Jun Ohya, Syunichi Yonemura, Makoto Enomoto

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • Image-based Assistance to Fire Extinguishing System with Mote

    Rubaiyat Yasmin, Jun Ohya, Aoki Yoshimitsu

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • 顔認識における二重符号化理論とデフォルメ似顔絵との関連性について - コンピュータによる似顔絵生成への応用 -

    榎本 誠, 大谷 淳, 楜沢 順

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • Motion Detection Based on Active Stereo Camera Array

    Xingdi Xie, Jun Ohya

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • 関節構造を持つ3次元弾性物体モデルの提案

    渡辺隆史, 大谷 淳, 楜沢 順, 徳永幸生

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • 画像処理を用いた店舗内の顧客の行動認識法に関する基礎的検討

    堀 隆之, 大谷 淳, 楜沢 順

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • センサーネットワークと監視カメラを用いた消火活動支援システムの提案

    野飼雅弘, Yasmin Rubaiyat, 大谷 淳

    FIT2005(第4回情報科学技術フォーラム) 

    Presentation date: 2005.09

  • A Proposal on Active Extensible Stereo Camera Array for High Speed Moving Object Detection

    Yingdi Xie, Jun Ohya

    情報処理学会オーディオビジュアル複合情報処理研究会 AVM2005 

    Presentation date: 2005.07

  • Analysis-based Participant Selection Scheme in Workflow Management System

    Ding Xiao, Cao Jian, Jun Ohya

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • A Study on MOTES to Assist Fire Extinguishing System

    Rubaiyat Yasmin, Jun Ohya, Aoki Yoshimitsu

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • A Fundamental Study of Topic Changes Prompted by the "●" Communication System

    Li Jen Chen, Yukio Tokunaga, Shunichi Yonemura, Jun Ohya

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • A Study of Synthesizing New Human Motions from Sampled Motions Using Tensor Decomposition

    Kalanov Rovshan, Jieun Cho, Jun Ohya

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • 因子分解法を用いた屋外環境下での広範囲における強化現実端末の自己位置・姿勢検出精度向上法の基礎的検討

    藤原裕也, 大谷淳, 野飼雅弘

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • 逆運動学による弾性物体の多関節モデルの検討

    渡辺隆史, 大谷淳, 楜沢順, 徳永幸生

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • 色彩画像処理による人物の3 次元姿勢推定に関する検討

    小沼裕一, 姜東完, 大谷淳

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • Shape from Shadingと遺伝的アルゴリズムを用いた水面形状推定法の基礎的検討

    寺島哲平, 大谷淳, 楜沢順, 徳永幸生

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • サイバーアクターの演技のタイミングに関する検討

    曹智恩, 大谷淳

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • サイバーシアターにおけるアヴァター表現に関する検討 - 画像処理による人物姿勢推定結果のニューラルネットによる改良-

    飯田将也, 姜東完, 大谷淳

    2005年電子情報通信学会総合大会 

    Presentation date: 2005.03

  • ソフトエッジとハードエッジに着目したコンピュータによる絵画作成法

    新原雄介, 徳永幸生, 楜沢順, 大谷淳

    情報処理学会第67回全国大会 

    Presentation date: 2005.03

  • 顔のエッジ表現を用いたコミュニケーションシステムの会話特性

    大石貴也, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会第67回全国大会 

    Presentation date: 2005.03

  • A Fundamental Analysis of Topic Changes Prompted by the "●"

    Li Jen Chen, Jun Ohya, Yukio Tokunaga, Tetsuya Kato

    電子情報通信学会技術研究報告,ヒューマンコミュニケーション基礎 

    Presentation date: 2005.03

  • 多関節化による弾性物体のインタラクティブモデルの提案

    渡辺隆史, 大谷淳, 楜沢順, 徳永幸生

    電子情報通信学会技術研究報告,マルチメディア・仮想環境基礎 

    Presentation date: 2005.03

  • ソフトエッジとハードエッジに着目したコンピュータによる絵画制作とその評価

    新原雄介, 徳永幸生, 楜沢順, 大谷淳

    電子情報通信学会技術研究報告,画像工学研究会 

    Presentation date: 2005.03

  • 顔のエッジ表現を用いたコミュニケーションシステム

    大石貴也, 徳永幸生, 米村俊一, 大谷淳

    情報処理学会研究会報告,ヒューマンインタフェース 

    Presentation date: 2005.01

  • Advanced Fire Extinguishing System: An Application of RFID With Ad hoc Networking

    Rubaiyat Yasmin, Jun Ohya, Aoki Yoshimitsu

    第2回QoSワークショップ予稿集 

    Presentation date: 2004.12

  • 熱転写記録を用いた金色銀色表現法とその応用

    徳永幸生, 古田裕樹, 楜沢順, 大谷淳

    パソコンリテラシ 

    Presentation date: 2004.10

  • Studies of the "●" Communication System

    Li Jen Chen, Tetsuya Kato, Yukio Tokunaga, Jun Ohya

    ヒューマンインタフェースシンポジウム2004 

    Presentation date: 2004.10

  • Research on expressing method of audience in a cyber-theater

    Kay Huang, Jun Ohya, Dong Wan Kang

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • LOCALIZATION OF HUMANOID ROBOT FROM MONOCULAR IMAGE SEQUENCE USING TWO LARGE TEMPLATE MATCHING

    Ni Ni Soe, Kenji Suzuki, Shuji Hashimoto, Jun Ohya

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • A Frame Work for Analysis and Synthesis of Actor's Stylistic Action

    Jieun Cho, Jun Ohya

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • 視覚次元互動システム:対話的な展示支援システムに関する提案 Visual-dimension Interact System(VIS)

    恩田淳, Chiu Pei-Yi, 奥智行, 横井真紀, 長幾朗, 大谷淳

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • Fundamentals of RFID and its Application Analysis in Smoky Area

    Rubaiyat Yasmin, Koji Koike, Aoki Yoshimitsu, Jun Ohya, Takagi Mikio

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • 仮想弾性物体の運動の操作手法に関する検討

    渡辺隆史, 大谷淳

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • A Communication System using a black prompter “●”

    Li Jen Chen, Tetsuya Kato, Yukio Tokunaga, Jun Ohya

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • 色彩画像処理と時系列情報による人物の3 次元姿勢推定に関する検討

    小沼裕一, 姜東完, 大谷淳

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • 屋外環境下での広範囲における強化現実端末の自己位置・姿勢検出精度向上法の基礎的検討

    藤原裕也, 大谷淳

    FIT2004 (第3回情報科学技術フォーラム) 

    Presentation date: 2004.09

  • Shape from Shading と