Details of a Researcher - WATANABE, Hiroshi

写真a

WATANABE, Hiroshi

Scopus Paper Info

Paper Count: 116 Citation Count: 508 h-index: 8

Click to view the Scopus page. The data was downloaded from Scopus API in April 25, 2024, via http://api.elsevier.com and http://www.scopus.com .

Google Scholar Information (Citations per year)

Citation Count: 4529 h-index: 22 i10-index: 38

Click to view the Google Scholar page.

Scopus Information

Affiliation

Faculty of Science and Engineering, School of Fundamental Science and Engineering

Job title

Professor

Degree

Doctor of Engineering ( Hokkaido University )

Homepage URL

https://www.ams.giti.waseda.ac.jp

Profile

Hiroshi Watanabe received his B.E., M.E. and Ph.D. degrees from Hokkaido University, Japan, in 1980, 1982 and 1985, respectively. He joined Nippon Telegraph and Telephone Corporation (NTT) in 1985, and was engaged in the research and development of image and video coding system at NTT Human Interface Labs. and NTT Cyber Space Labs. until 2000. He has been involved with the development of JPEG, MPEG standards under JTC 1/SC 29. In 2000, He moved to Waseda University as a Professor at the Graduate School of Global Information and Telecommunication Studies. He is currently a professor at the Department of Communications and Computer Engineering, School of Fundamental Science and Engineering. He served as an ISO/IEC JTC 1/SC 29 Chairman from November 1999 to 2006. He was an associate editor of IEEE Transactions of Circuits and Systems for Video Technology as well as Communications. His research interests include image and video processing, video coding, and multimedia distribution. He is a member of a steering committee of PCSJ. He is a member of IEEE, IEICE, IPSJ, ITE and IIEEJ.

Research Experience

2013.04

-

　

Waseda University, School of Fundamental Science and Engineering Dept. of Communications and Computer Engineering Professor
2001.04

-

2013.03

Waseda University, Graduate School of Global Information and Telecommuniaction Studies Professor
2000.09

-

2001.03

Waseda University, Global Information and Telecommunication Institute Professor
1995.04

-

2000.09

NTT Human Interface Laboratories Senior Research Engineer, Supervisor
1991.03

-

1994.03

NTT Human Interface Laboratories, Senior Distinguished Senior Research Engineer
1990.06

-

1991.05

Bell Communications Research Visiting Researcher
1987.04

-

1989.03

NTT Human Interface Laboratories Senior Research Engineer
1987.04

-

1989.03

NTT Human Interface Laboratories Dept. of Visual Communication Research Engineer
1985.04

-

1987.03

NTT Yokosuka Electrical Communication Dept. of Visual Communication Researcher

▼display all

Education Background

1980.04

-

1985.03

Hokkaido University Graduate School, Faculty of Engineering Electronic Engineering
1976.04

-

1980.03

Hokkaido University Faculty of Engineering Electronic Engineering

Professional Memberships

　

　

　

IEEE (The Institute of Electrical and Electronic Engineers)
　

　

　

情報処理学会
　

　

　

画像符号化シンポジウム
　

　

　

IEEE Circuits and Systems Society
　

　

　

電子情報通信学会
　

　

　

画像電子学会
　

　

　

映像情報メディア学会

▼display all

Research Areas

Database

Research Interests

Information technology, Communication Engineering

Awards

情報処理学会情報規格調査会標準化貢献賞

2005
日本規格協会標準化文献賞奨励賞

1998
画像電子学会技術賞

1997
映像情報メディア学会丹羽高柳著述賞

1997
電子情報通信学会学術奨励賞

1988

Papers

A Real-time Polyp Detection Method Based on GhostAtt-YOLOv8

Phong Thao Nguyen, Hiroshi Watanabe

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) 2024.03 [Refereed]
Object Detection Method for Drone Videos Using Optical Flow

Takahiro Shindo, Kein Yamada, Taiju Watanabe, Hiroshi Watanabe

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) ( LBP-05 ) 2024.03
Image Coding for Machines with Objectness-based Feature Distillation

Kei Iino, Shunsuke Akamatsu, Hiroshi Watanabe, Shohei Enomoto, Akira Sakamoto, Takeharu Eda

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) ( LBP-07 ) 2024.03
Edge-Cloud Collaborative Object Detection Model with Feature Compression

Shunsuke Akamatsu, Kei Iino, Hiroshi Watanabe, Shohei Enomoto, Akira Sakamoto, Takeharu Eda

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) ( LBP-12 ) 2024.03
3D Pose Estimation Using Time Series Data in Event-based Video

Kakeru Koizumi, Hiroshi Watanabe

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) ( LBP-28 ) 2024.03
Data Augmentation with 3D-rendered Models for Livestock Recognition Using Drone Footage

Kein Yamada, Takahiro Shindo, Taiju Watanabe, Hiroshi Watanabe

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) ( BP-29 ) 2024.03
The Effect of Edge Information in Stable Diffusion Applied to Image Coding

Hiroshi Watanabe, Luoxu Jin, Taiga Hayami, Takeshi Chujoh, Yukinobu Yasugi, Sujun Hong, Zheming Fan, Tomohiro Ikai

IIEEJ International Conference on Image Electronics and Visual Computing (IEVC2024) ( LBP-15, ) 2024.03
Introduction of Auxiliary Loss in Deep Image Compression for Image Recognition

Kei Iino, Shunsuke Akamatsu, Hiroshi Watanabe, Takeharu Eda, Shohei Enomoto, Akira Sakamoto

IEICE Genaral Conference ( D12A-37 ) 2024.03
A Study on Point Cloud Classification Method Integrating Residual MLP and Attention Mechanism

Hongui Fan, Hiroshi Watanabe

IEICE General Conference ( D-12B-13 ) 2024.03
Quality Improvement of NeRF-based Frame Interpolation Method

Taiga Hayami, Luoxu Jin, Hiroshi Watanabe

IEICE General Conference ( D-11A-26 ) 2024.03
Prompt-based Image Coding with Edge and Color Information

Hiroshi Watanabe, Luoxu Jin, Taiga Hayami, Takeshi Chujoh, Yukinobu Yasugi, Sujun Hong, Zheming Fan, Tomohiro Ikai

IEICE General Conference ( D-11A-27 ) 2024.03
Image Coding for Machines with Object Region Learning

Takahiro Shindo, Taiju Watanabe, Kein Yamada, Hiroshi Watanabe

IEEE Consumer Communications & Networking Conference (CCNC 2024) 2024.01 [Refereed]
Enhancing Gaze Estimation through Fine-Grained Analysis of Eye Region

Hideharu Sugiyama, Hiroshi Watanabe

International Workshop on Advanced Image Technology (IWAIT 2024) ( No. 11 ) 2024.01 [Refereed]
Attention-Enhanced Graph Convolutional Network for Assessing Rehabilitation Exercises

Smita Priyadarshani, Hiroshi Watanabe

International Workshop on Advanced Image Technology (IWAIT 2024) ( Vol. 73 ) 2024.01 [Refereed]
Prompt-based Image Coding with Edge Information

Hiroshi Watanabe, Luoxu Jin, Taiga Hayami, Takeshi Chujoh, Tomoko Aono, Yukinobu Yasugi, Sujun Hong, Zheming Fan, Tomohiro Ikai

2023 Picture Coding Symposium, Image Media Symposium (PCSJ/IMPS2023) ( P1-12 ) 2023.11
Characterization of frame interpolation methods based on NeRF and feature maps

Taiga Hayami, Luoxu Jin, Hiroshi Watanabe, Takeshi Chujoh, Tomoko Aono, Yukinobu Yasugi, Sujun Hong, Zheming Fan, Tomohiro Ikai

2023 Picture Coding Symposium, Image Media Symposium (PCSJ/IMPS2023) ( P2-05 ) 2023.11
Post-processing Based Image Coding via Stable Diffusion

Luoxu Jin, Hiroshi Watanabe, Taiga Hayami, Takeshi Chujoh, Tomoko Aono, Yukinobu Yasugi, Sujun Hong, Zheming Fan, Tomohiro Ikai

2023 Picture Coding Symposium, Image Media Symposium (PCSJ/IMPS2023) ( P3-08 ) 2023.11
Video Salient Object Detection Using Multi-Scale Self-Attention

Jiahao Liu, Haoyuan Liu, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-VDP ) 370 - 373 2023.11 [Refereed]
BigWavGAN: A Wave-To-Wave Generative Adversarial Network for Music Super-Resolution

Yenan Zhang, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-AAT ) 596 - 6000 2023.11 [Refereed]
Future Object Detection Using Frame Prediction

Taiju Watanabe, Takahiro Shindo, Kein Yamada, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-TMR ) 948 - 949 2023.11 [Refereed]
VVC Extension Scheme for Object Detection Using Contrast Reduction

Takahiro Shindo, Taiju Watanabe, Kein Yamada, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-TMR ) 1104 - 1105 2023.11 [Refereed]
Accuracy Consistency of Object Detection With Contrast Reduction by Pixel Value Limitation

Kein Yamada, Takahiro Shindo, Taiju Watanabe, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-TMR ) 1108 - 1111 2023.11 [Refereed]
Point Cloud Denoising and Outlier Detection with Local Geometric Structure by Dynamic Graph CNN

Kosuke Nakayama, Hiroto Fukuta, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-AIP ) 156 - 159 2023.10 [Refereed]
Three-Dimensional Baseball Strike Judgement by Monocular Video from Umpire’s Viewpoint

Hiroto Fukuta, Kosuke Nakayama, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-AIP ) 160 - 163 2023.10 [Refereed]
Feature Transfer Block for Feature Fusion in Lightweight Object Detectors

Haoyuan Liu, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-AIR ) 304 - 307 2023.10 [Refereed]
A Video Object Detection Method of ECNet Based on Frame Difference and Grid Cell Confidence

Shunsuke Akamatsu, Kei Iino, Hiroshi Watanabe, Shohei Enomoto, Xu Shi, Akira Sakamoto, Takeharu Eda

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-VDP ) 366 - 369 2023.10 [Refereed]
Measure the Difference Between Frames: Reconsidering the Evaluation Method in Dynamic Mesh Simplification

Yunlong Liu, Jianfeng Xu, Kei Kawamura, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2023) ( OS-VDP ) 374 - 377 2023.10 [Refereed]
Accuracy Improvement of Object Detection in VVC Coded Video Using YOLO-v7 Features

Takahiro Shindo, Taiju Watanabe, Kein Yamada, Hiroshi Watanabe

IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET2023) ( No. 139 ) 1 - 5 2023.09 [Refereed]
A Study on Frame Reduction Methods to Reduce the Amount of Processing for Skelton-Based Motion Matching

Shohei Adachi, Hiroshi Watanabe

IEICE General Conference ( D-12-42 ) 2023.03
A Study of Introducing Attention Module into Feature Compression Model

Kei Iino, Miho Takahashi, Hiroshi Watanabe, Takeharu Eda, Shohei Enomoto, Akira Sakamoto, Xu Shi, Ichiro Morinaga

IEICE General Conference ( D-11-12 ) 2023.03
Video Coding Scheme for YOLO-v7 Combining VVC and CNN

Takahiro Shindo, Taiju Watanabe, Hiroshi Watanabe

IEICE General Conference ( D-11-9 ) 2023.03
A Study of Appearance-based Gaze Estimation Models with Eye Features

Hideharu Sugiyama, Hiroshi Watanabe

IEICE General Conference ( D-12-52 ) 2023.03
Outlier Detection for Point Cloud with Local Geometric Structure

Kosuke Nakayama, Junichi Hara, Hiroshi Watanabe

IEICE General Conference ( D-12-35 ) 2023.03
A Study on 3D Pose Estimation with Absolute Skeletal Length

Hiroto Fukuta, Sho Nakajima, Hiroshi Watanabe

IEICE General Conference ( D-12-67 ) 2023.03
A Study on Super-Resolution Network Using Partitioned Block of Coded Video

Remina Yano, Yun Liu, Hiroshi Watanabe, Takuya Suzuki, Zheming Fan, Takeshi Chujoh, Tomohiro Ikai

IEICE General Conference ( D-11-10 ) 2023.03
A Study on Future Object Detection Using YOLOV

Taiju Watanabe, Takahiro Shindo, Hiroshi Watanabe

IEICE General Conference ( D-11-11 ) 2023.03
Research on Video Captioning with a Late Fusion Based Multimodal Transformer Network

Fei Bao, Hiroshi Watanabe

IPSJ National Convension ( IQ-08 ) 2-195 - 2-196 2023.03
A Study on Raindrop Removal Method Using GAN and Templete Matching

Marica Arimoto, Junichi Hara, Hiroshi Watanabe

IPSJ-AVM Vol.2023-AVM-120 ( No.3 ) 1 - 4 2023.03
A Study on 3D Pose Estimation Method with Skeltal Symmetry

Sho Nakajima, Hiroto Fukuta, Hiroshi Watanabe

IPSJ-AVM AVM-120 ( No. 4 ) 1 - 2 2023.03
A Study on Background Filling for the Inter-patch Area of the Texture Image in V-PCC Encoder

Shohei Adachi, Yoshitaka Kidani, Kei Kawamura, Hiroshi Watanabe

IEICE Technical Report of Image Engineering (IE) ( IE2022-54 ) 17 - 18 2023.02 [Invited]
Category-based Memory Bank Design for Traffic Surveillance in Context R-CNN

Miho Takahashi, Kei Iino, Hiroshi Watanabe, Ichiro Morinaga, Shohei Enomoto, Xu Shi, Akira Sakamoto, Takeharu Eda

International Workshop on Advanced Image Technology 2023 (IWAIT 2023) ( Vol.12592 ) 125920G-1 - 125920G-4 2023.01 [Refereed]

DOI

Scopus
Novel CNN Approach for Video Prediction Based on FitVid

Taiju Watanabe, Takahiro Shindo, Hiroshi Watanabe

Proceedings of SPIE for IWAT2023 ( Vol. 12592 ) 1259210-1 - 1259210-4 2023.01 [Refereed]

DOI

Scopus
Memory Bandwidth Constrained Overlapped Block Motion Compensation for Video Coding

Yoshitaka Kidani, Kyohei Ueno, Kei Kawamura, Hiroshi Watanabe

ITE Transactions on Media Technology and Applications (MTA), Vol. 11 ( No. 1 ) 1 - 12 2023.01 [Refereed]

DOI

Scopus

1

Citation

(Scopus)
A Prior-Guided Face Image Super-Resolution Network Based on Attention Mechanism

Yun Liu, Remina Yano, Hiroshi Watanabe, Takuya Suzuki, Zheming Fan, Takeshi Chujoh, Tomohiro Ikai

Proceedings of SPIE - The International Society for Optical Engineering 12592 ( No. 7 ) 2023 [Refereed]

　View Summary

In this paper, we propose a face image super-resolution method, aiming to reconstruct high quality face images from a low resolution input. The proposed method introduces an attentional multi-scale feature fusion block, which aims to improve the representation power of the neural network by emphasizing the important feature maps and suppressing the unimportant ones. In addition, the facial prior information is utilized by adding a separate prior branch, an hourglass structure is used. Experiments show that the face images reconstructed by the proposed method exhibit noticeable quality improvement compared to the low-resolution images and other SR approaches.

DOI

Scopus
Application of Multi-modal Fusion Attention Mechanism in Semantic Segmentation

Yunlong Liu, Osamu Yoshie, Hiroshi Watanabe

Proceedings of the Asian Conference on Computer Vision (ACCV2022) 1245 - 1264 2022.11 [Refereed]

DOI

Scopus
Super Resolution for QR Code Images

Takahiro Shindo, Taiju Watanabe, Remina Yano, Marika Arimoto, Miho Takahashi, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2022) ( OS-TMR(2 ) 274 - 277 2022.10 [Refereed]

DOI
A Method for Improving Object Detection Accuracy in Coding Noise Environment

Takahiro Shindo, Taiju Watanabe, Hiroshi Watanabe

PCSJ/IMPS2022 ( P1-01 ) 14 - 15 2022.10
A Study on Background Filling of Inter-patch Area of Texture Image In V-PCC Encoder

Shohei Adachi, Yoshitaka Kidani, Kei Kawamura, Hiroshi Watanabe

PCSJ/IMPS2022 ( P2-09 ) 72 - 73 2022.10
Selective Use of Skeletal Information to Reduce Computational Complexity of Motion Matching

Shohei Adachi, Ryohei Osawa, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2022) ( OS-ADE ) 68 - 71 2022.10 [Refereed]

DOI

Scopus
Motion Matching Based on Pose Similarity Using Movement of Body Parts

Ryohei Osawa, Shohei Adachi, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2022) ( OS-ADE ) 83 - 86 2022.10 [Refereed]

DOI

Scopus
Inter-Feature-Map Differential Coding of Surveillance Video

Kei Iino, Miho Takahashi, Hiroshi Watanabe, Ichiro Morinaga, Shohei Enomoto, Xu Shi, Akira Sakamoto, Takeharu Eda

IEEE Global Conference on Consumer Electronics (GCCE2022) ( OS-ASP(1) ) 293 - 296 2022.10 [Refereed]

DOI

Scopus

1

Citation

(Scopus)
Improving Accuracy of Position Information in Egocentric Pose Estimation Methods

Sho Nakashima, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2022) ( OS-DSC ) 688 - 691 2022.10 [Refereed]

DOI

Scopus
Geometric Partitioning Mode with Inter and Intra Prediction for Beyond Versatile Video Coding

EICE Transaction Information & Systems Vol. E105-D ( No.10 ) 1691 - 1703 2022.10 [Refereed]
Development of high-definition technology for microscopic images via the fusion of Super-Resolution technology applying CNN and denoising processing

Niai Yano, Tomoe Ishikawa, Hiroshi Watanabe, Masato Yasui

The 84th National Convention of IPSJ ( 6Q-06 ) 2-259 - 2-260 2022.03
Sports Motion Matching Based on Athelete’s Pose Similarity

AVM-116 ( No. 5 ) 1 - 3 2022.02
A Study on Anormaly Detection Method Using Normal Images

Yusuke Yamanoi, Junichi Hara, Hiroshi Watanabe

IPSJ Technical Report AVM-116, ( No.7 ) 1 - 4 2022.02
GAN Using Capsule Network for Discriminator and Generator

Kanako Marusaki, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2021) ( OS-AIP(2) ) 726 - 729 2021.10 [Refereed]
An Image Inpainting Method Considering Edge Connectivity of Defects

Marika Arimoto, Junichi Hara, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2021) ( OS-TMR ) 127 - 128 2021.10 [Refereed]
Face Image Generation for Illustration by WGAN-GP Using Landmark Information

Miho Takahashi, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2021) ( OS-DSC ) 989 - 990 2021.10 [Refereed]
Text Image Super Resolution Using Deep Attention Neural Network

Yun Liu, Remina Yano, Hiroshi Watanabe, Takuya Suzuki, Takeshi Chujoh, Tomohiro Ikai

IEEE Global Conference on Consumer Electronics (GCCE2021) ( OS-TMR(3) ) 306 - 308 2021.10 [Refereed]
Variant Graph Convolutional Networks for Skeleton-Based Hand Action Recognition

Khin Sabai Htwe, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2021) ( OS-AIP(2) ) 730 - 731 2021.10 [Refereed]
Multi Frame Super-Resolution for Versatile Video Coding

Toshiya Hori, Remina Yano, Hiroshi Watanabe, Takeshi Chujoh, Tomohiro Ikai, Eiichi Sasaki, Takuya Suzuki, Norio Ito

SPIE Procedings for IWAIT2021 ( 11766 ) 1 - 6 2021.04 [Refereed]

DOI

Scopus
Creation of 3D Environmental Map using Omnidirectional Camera Images

Yuya Kato, Junichi Hara, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2021) ( OS-VDP ) 522 - 525 2021.03 [Refereed]
Generation and Extraction of Color Palettes with Adversarial Variational Auto Encoders

Ahmad Moussa, Hiroshi Watanabe

6th International Congress on Information and Communication Technology (ICICT2021) 1 - 8 2021.02 [Refereed]
Multi Frame Super-Resolution for Versatile Video Coding

Toshiya Hori, Remina Yano, Hiroshi Watanabe, Takeshi Chujoh, Tomohiro Ikai, Eiichi Sasaki, Takuya Suzuki, Norio Ito

International Workshop on Advanced Image Technology (IWAIT2021) ( 45 ) 2021.01 [Refereed]
Coded Video Transmission using Super-Resolution and MPEG-NNR Deep Learning Metadata

Takuya Suzuki, Eichi Sasaki, Takeshi Chujoh, Tomohiro Ikai, Hiroshi Watanabe

2020 Picture Coding Symposium Japan ( P3-B-1 ) 2020.11
Pitching Motion Matching Based on Pose Similarity Using Dynamic Time Warping

Ryohei Osawa, Takaaki Ishikawa, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 140 - 144 2020.09 [Refereed]
Pseudo Ground Truth Segmentation Mask to Improve Video Prediction Quality

Muchien Hsu, Juichun Shyur, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 151 - 152 2020.09 [Refereed]
Facial Age Estimation by Curriculum Learning

Wei Wang, Takaaki Ishikawa, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 203 - 204 2020.09 [Refereed]
ECNet: A Fast, Accurate, and Lightweight Edge-Cloud Network System Based on Cascading Structure

Libo Hu, Tao Wang, Hiroshi Watanabe, Shohei Enomoto, Xu Shi, Akira Sakamoto, Takeharu Eda

IEEE Global Conference on Consumer Electronics (GCCE2020) 259 - 262 2020.09 [Refereed]
Animal Behavior Classification Using DeepLabCut

Shiori Fujimori, Takaaki Ishikawa, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 259 - 262 2020.09 [Refereed]
Study on Improvement of Estimation Accuracy in Pose Estimation Model Using Time Series Correlation

Atsuya Yamakawa, Takaaki Ishikawa, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 570 - 573 2020.09 [Refereed]
Rollover Detection of Infants Using Posture Estimation Mode

Ayaka Okuno, Takaaki Ishikawa, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 688 - 691 2020.09 [Refereed]
Self-Attention Based Neural Network for Few Shot Classification

Ravi Jain, Hiroshi Watanabe

IEEE Global Conference on Consumer Electronics (GCCE2020) 739 - 740 2020.09 [Refereed]
Transfer Rate Estimation in Edge-Cloud Neural Network Solution for Object Detection

Libo Hu, Tao Wang, Yucheng Zhou, Hiroshi Watanabe, Shohei Enomoto, Xu Shi, Akira Sakamoto, Takeharu Eda

IEICE General Conference ( D-11-20 ) 2020.03
High-Resolution Processing Applying Machine Learning Based Multi-Step Super Resolution

Niai Yano, Hiroshi Watanabe

IEICE General Conference ( D-11-23 ) 2020.03
PSNR Adaptive Super-resolution for Images with Coding Noise

Toshiya Hori, Zichen Gong, Seiya Umeda, Hiroshi Watanabe, Takeshi Chujo, Tomohiro Ikai, Eichi Sasaki, Norio Ito

IEICE General Conference ( D-11-33 ) 2020.03
Exit-Point Setting in Edge-Cloud Solution for Object Detection

Tao Wang, Libo Hu, Hiroshi Watanabe, Shohei Enomoto, Xu Shi, Arkira Sakamoto, Takeharu Eda

IEICE General Conference ( D-12-14, ) 2020.03
Expression Recognition Using Curriculum Learning Method

Wei Wang, Takaaki Ishikawa, Hiroshi Watanabe

IEICE General Conference ( D-12-49 ) 2020.03
An Automatic Music Arrangement System Using Machine Learning

Tianyi Gao, Hangyu Song, Ahmad Moussa, Hiroshi Watanabe

IEICE General Conference ( D-14-3 ) 2020.03
Improving Pose Estimation Models Using Time Series Correlation

Atsuya Yamakawa, Takaaki Ishikawa, Hiroshi Watanabe

National Convention of Information Processing Society of Japan ( 1R-05 ) 2020.03
A Study on Pitching Scene Matching Using OpenPose

Ryohei Osawa, Takaaki Ishikawa, Hiroshi Watanabe

National Convention of Information Processing Society of Japan ( 1R-06 ) 2020.03
Image Quality Enhancement with Machine Learning Based Multi-Step Super-Resolution

Niai Yano, Hiroshi Watanabe

IEEE The 2nd International Conference on Artificial Intelligence in Information and Communication (ICAIIC 2020) ( 3B-6 ) 1 - 5 2020.02 [Refereed]
Audio Translation with Conditional Generative Adversarial Networks

Ahmad Yunis Moussa, Hiroshi Watanabe

IEEE The 2nd International Conference on Artificial Intelligence in Information and Communication (ICAIIC 2020) ( 9B-1 ) 1 - 5 2020.02 [Refereed]
Automatic Classification of Manga Characters using Density-Based Clustering

H. Yanagisawa, K. Kyogoku, R. Jain, H. Watanbe

International Workshop on Advanced Image Technology, IWAIT 2020 ( No. 20 ) 1 - 6 2020.01 [Refereed]
CNN-based Super-Resolution Adapted to Quantization Parameters

T. Hori, Z. Gong, H. Watanabe, T. Ikai, T. Chujoh, E. Sasaki, N. Ito

International Workshop on Advanced Image Technology, IWAIT 2020 ( No. 42 ) 1 - 5 2020.01 [Refereed]
A Selective Fusion Module for Video Super Resolution with Recurrent Architecture

Z. Gong, T. Hori, H. Watanabe, T. Ikai, T. Chujoh, E. Sasaki, N. Ito

International Workshop on Advanced Image Technology, IWAIT 2020 ( No. 43 ) 1 - 6 2020.01 [Refereed]
A Study on Pitching Scene Matching Using OpenPose and Dynamic Time Warping

Ryohei Osawa, Takaaki Ishikawa, Hiroshi Watanabe

ITE Winter Annual Convention 2019 ( 11A3 ) 2019.12
A Study on Scene Matching for Different Pitching Styles

Ryohei Osawa, Takaaki Ishikawa, Hiroshi Watanabe

ITE Technical Report ME2019-117 ( Vol.43, No.39 ) 13 - 14 2019.11
A Study on Generative Adversarial Network Using Capsule Network for Generator

Kanako Marusaki, Hiroshi Watanabe

2019 Picture Coding Symposium・2019 Image Media Processing Symposium ( P-1-03 ) 2019.11
Auto Arrangement and Threshold to Database for Face Recognition

MuChien Hsu, Ju Chun Shyur, Hiroshi Watanabe

( P-1-11 ) 2019.11
Few Shot Classification by Learning to Learn

2019 Picture Coding Symposium・2019 Image Media Processing Symposium ( P-1-15 ) 2019.11
A Study on Generation of Japanese Sweets Images by Generative Adversarial Networks

Yuka Oyama, Hiroshi Watanabe

2019.11
A Study on 3D Reconstruction Using Omnidirectional Camera and Alignment Method

Yuya Kato, Junichi Hara, Hiroshi Watanabe

2019 Picture Coding Symposium・2019 Image Media Processing Symposium ( P-3-17 ) 2019.11
A Study about Classification of Manga Character Faces using Density-Based Clustering

Hideaki Yanagisawa, Kengo Kyogoku, Ravi Jain, Hiroshi Watanabe

2019 Picture Coding Symposium・2019 Image Media Processing Symposium ( P-4-07 ) 2019.11
Two-side Network for Person Detection and Person Re-identification

Libo Hu, Tao Wang, Hiroshi Watanabe

2019 Picture Coding Symposium・2019 Image Media Processing Symposium ( P-4-12 ) 2019.11
A Study on Super-resolution of Noisy Images by Scaling and Coding

A Study on Super-resolution of Noisy Images by Scaling and Coding

PCSJ/IMPS2019 ( P-4-12 ) 2019.11
Improvement of Feature Representation in Clustering of Manga Character Faces

H. Yanagisawa, K. Kyogoku, R. Jain, H. Watanabe

FIT2019 H-029 2019.09
Land Cover Classification and Change Detection Analysis of Multispectral Satellite Images Using Machine Learning

SPIE Remote Sensing 2019, Image and Signal Processing for Remote Sensing XXV 11155-56 2019.09
Hand Joints Detection on Noisy Hand Poses Using Variational Autoencoder

Khin Sabai Htwe, Takaaki Ishikawa, Hiroshi Watanabe

The 6th IIEEJ International Conference on Image Electronics and Visual Computing (IEVC 2019) 4B-2 2019.08
A Study on 3D Reconstruction from Omnidirectional Movie and Registration Method

Y. Kato, J. Hawa, H. Watanabe

ITE Annual Conference 34B-3 2019.08
Image Quality Improvement of Sequential Images by Different Combination Orders of Learning-based and Reconstruction-based Super Resolution

T. Hori, Y. Kato, H. Watanabe

ITE Annual Conference 32B-1 2019.08
A Study on Sport Scene Matching

R. Osawa, T. Ishikawa, H. Watanabe

ITE Annual Conference 31C-3 2019.08
Deep Learning Based Speaker Recognition System

ITE Annual Conference 13B-1 2019.08
Analyzing Hand Action Recognition using Spatial Temporal Graph Convolutional Networks (ST-GCN)

ITE Annual Conference 11B-3 2019.08
Application of Omnidirectional Video to LSD-SLAM by Image Conversion and Alignment

Y. Kato, J. Hara, H. Watanabe

情報処理学会オーディオ・ビジュアル複合情報処理（AVM）研究会 AVM105, No.14, pp.1-4 2019.06
Feature Extraction of Point Cloud for Noise Shaping of Point Cloud

H. Akatsuka, J. Hara, H. Watanabe

情報処理学会オーディオ・ビジュアル複合情報処理（AVM）研究会 AVM105, No.13, pp.1-4 ( 16 ) 73 - 75 2019.06

CiNii
Classification of Signal Japanese Character with Pose Estimation and Machine Learning

H. Kobayashi, T. Ishikawa, H. Watanabe

IEICE General Conference D-12-62 2019.03
Basic Study on Lip Reading for Japanese Speaker by Machine Learning

R. Asami, T. Ishikawa, H. Watanabe

IEICE General Conference D-12-61 2019.03
Creation of Reference Model of Multiple Dancers Using Machine Learning

K. Inada, T. Ishikawa, H. Watanabe

IEICE General Conference D-12-60 2019.03
Alignment for Skeleton Coordinated Data Obtained from Sports Video

S. Yokoi, T. Ishikawa, H. Watanabe

IEICE General Conference D-12-59 2019.03
Generation of Wagashi Images by GAN and Its Evaluation

Y. Oyama, H. Watanabe

IEICE General Conference D-12-40 2019.03
A Study on Feature Extraction of Point Cloud

H. Akatsuka, J. Hara, H. Watanabe

IEICE General Conference D-12-39 2019.03
Detection of Corresponding Lines from Omnidirectional Images

T. Aoki, J. Hara, H. Watanabe

IEICE General Conference D-12-36 2019.03
3D Reconstruction from Omnidirectional Video Using LSD-SLAM

Y. Kato, J. Hara, H. Watanabe

IEICE General Conference D-12-35 2019.03
Sign Language Analysis with Monocular RGB Camera

T. Yanagisawa, T. Ishikawa, H. Wataneb

IEICE General Conference D-12-34 2019.03
Content-Independent Speaker Recognition System based on Neural Networks

H. Song, H. Watanabe

IEICE General Conference D-12-96 2019.03
A Study on GAN Using Capsule Network

K. Marusaki, H. Watanabe

IEICE General Conference D-12-8 2019.03
Characteristics of Bit Depth Extention by Super Resolution

S. Umeda, H. Watanabe, T. Ikai, T. Hashimoto, T. Chujo, N. Ito

IEICE General Conference D-12-7 2019.03
Image Quality Enhancing by Multi-Step Super-Resolution

N. Yano, H. Watanabe

IEICE General Conference D-11-9 2019.03
Quality Improvement of Sequential Images Using Super Resolution

T. Hori, H. Watanabe

IEICE General Conference D-11-6 2019.03
Comparison of Machine Learning Classifiers for Land Cover Mapping in Google Earth Engine,

Nyein Soe Thwal, T. Ishikawa, H. Watanabe

IEICE General Conference BS-4-21 2019.03
An Evaluation of The Impact of Dataset Bias in Pretrained VGG Network on The Performance of Neural Network Based Style Transfer

Z. Gong, H. Watanabe

IEICE General Conference BS-4-19 2019.03
Complex Character Retrieval from Comics using Deep Learning

R. Jain, H. Watanabe

IEICE General Conference BS-4-18 2019.03
Study on Feature Extraction of Point Cloud Using Principal Component Analysis

H. Akatsuka, J. Hara, H. Watanabe

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 AVM104-003 2019.02
スポーツ情報処理と映像表現への期待

渡辺裕

NHK技研 R&D No.173, pp.2-3 2019.01
Clustering of Comic Character Images for Exraction of Major Characters

H. Yanagisawa, T. Yamashita, H. Watanabe

ITE Online Journal Vol. 73, No.1, pp.199-204 2019.01
Automatic Self-Improvement Scheme in Optical Flow-Based Motion Estimation for Sequential Fisheye Images

Arief Suryadi Satyawan, Junichi Hara, Hiroshi Watanabe

ITE Transactions on Media Technology and Applications Vol. 7, No. 1, pp.20-35 ( 1 ) 20 - 35 2019.01

　View Summary

This paper aims to present an innovative design of motion estimation for sequential fisheye images. This design is an extended version of the original Lucas and Kanade's (LK) concept that used to design for calculating optical flow from general perspective images. The extended design consists of the LK concept and an additional self-improvement mechanism that automatically finds the maximum performance of the estimated motion. This extended scheme works much better than the original LK's idea or some block-based motion estimations. Moreover, to some extent, this proposed method is working extremely well to overcome some critical characteristics of the sequential fisheye images. These characteristics include distortion error on the fisheye image area, inconsistent brightness level, fluctuating number of object motion, changing the shape of object motion, or poor camera stability.

DOI

Scopus

2

Citation

(Scopus)
Manga Character Clustering with DBSCAN Using Finetuned CNN Model

H. Yanagisawa, T. Yamashita, H. Watanabe

IWAIT-FMIA 2019 No.202 2019.01
Joint Super-Resolution and Bit Depth Extention by DNN

S. Umeda, H. Watanabe, T. Ikai, T. Hashimoto, T. Chujo, N. Ito

IWAIT-FMIA 2019 No.201 2019.01

　View Summary

Many Super-Resolution algorithm using DNN have been proposed. They simply expand the image size. On the other hand, bit depth expansion has also been proposed independently. In this paper, we consider the case when the image enlargement process is combined with bit depth expansion by Super-Resolution processing. Experimental results show its validity in both SDR and HDR images.

DOI

Scopus

1

Citation

(Scopus)
Sequence Alignment of Time-Series Multidimensional Data Extracted from Sports Video

S. Yokoi, T. Ishikawa, H. Watanabe

ITE SIP Technical Report ME2018-115 (SIP2018-03-12), pp.43-46 2018.12
縦方向の回転運動に対するOpenPose”

S. Kawashima, T. Ishikawa, H. Watanabe

ITE SIP Technical Report ME2018-110 (SIP2018-03-07), pp.25-28 2018.12
OpenPoseを用いた複数ダンサーの姿態評価の基礎検討

K. Inada, T. Ishikawa, H. Watanabe

ITE SIP Technical Report ME2018-109 (SIP2018-03-06), pp.21-24 2018.12
Mapping Urban Land Cover by Using Landsat Sentinel Images and Open Social Data

Nyein Soe Thwal, Takaaki Ishikawa, Hiroshi Watanabe

ITE Winter Annual Convention 22D-4 2018.12
Visual Improvement of Image Quality by Multiple-Step Super-Resolution

N. Yano, H. Watanabe

ITE Winter Annual Convention 14D-5 2018.12
On Quality Improvement of Super Resolution Processing for Coded Images

S. Umeda, N. Yano, H. Watanabe

ITE Winter Annual Convention 14D-4 2018.12
Pose Estimation of Rotation Invariant Motion Using OpenPose

S. Kawashima, T. Ishikawa, H. Watanabe

ITE Winter Annual Convention 13D-3 2018.12
Evaluation of Action Synchronization and Shape for Multiple Dancers Using OpenPose

K. Inada, T. Ishikawa, H. Watanabe

ITE Winter Annual Convention 13D-2 2018.12
Event Recognition in SPorts Video Using Motion Similarity

S. Yokoi, T. Ishikawa, H. Watanabe

ITE Winter Annual Convention 13D-1 2018.12
Improving Detection of Hand Joints in RGB Images using Maximum Confidence Value

Khin Sabai Htwe, Takaaki Ishikawa, Hiroshi Watanabe

ITE Winter Annual Convention 12D-5 2018.12
Data Augmentation Using CapsGAN Considering Generated Image Quality

T. Matsuda, K. Marusaki, H. Watanabe

ITE Winter Annual Convention 12D-4 2018.12
Comparison of CNN based Illustration Drawing-Style Classification Systems

Hangyu Song, Hiroshi Watanabe

ITE Winter Annual Convention 12D-3 2018.12
Noise Shaping of 3D Point Clouds Using Omni Image Features

H. Akatsuka, J. Hara, H. Watanabe

ITE Winter Annual Convention 11D-5 2018.12
Simultaneous Localization and Mapping Using Omnidirectional Camera

T. Aoki, H. Watanabe, J. Hara

ITE Winter Annual Convention 11D-4 2018.12
Basic Study of Evaluation of Body Shape for Multiple Dancer Using OpenPose

K. Inada, T. Ishikawa, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-5-7 2018.11
Super-Resolution Technology Using Multiple Output Images of Example-Based Super-Resolution

N. Yano, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-4-8 2018.11
全天球カメラを用いた自己位置推定と環境地図作成の基礎検討

青木, 渡辺, 原

2018年画像符号化シンポジウム・2018年映像メディア処理シンポジウム（PCSJ/IMPS2018) P-4-7 2018.11
Boundary Detection of 3D Point Clouds Using Omni Image Features

H. Akatsuka, J. Hara, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-3-9 2018.11
Basic Study on Lip Reading for Japanese Speaker by Machine Learning

R. Asami, T. Ishikawa, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-3-8 2018.11
Hand Joints Detection Using Initial Hand Detector on Egocentric Views

Khin Sabai Htwe, Takaaki Ishikawa, Hiroshi Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-3-3 2018.11
Event Recognition in Sports Video Using Body Part Coordinates

S. Yokoi, T. Ishikawa, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-2-10 2018.11
Rotation-Invariant Pose Detection Using OpenPose to Rotate Action

S. Kawashima, T. Ishikawa, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-2-7 2018.11
Land Cover Classification Using Landsat and Sentinel Images by Integrating Pixel and Object-based Segmentation Approach

Nyein Soe Thwal, Takaaki Ishikawa, Hiroshi Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-2-3 2018.11
Comparison of CNN based Illustration Drawing-Style Classification Systems

Hangyu Song, Hiroshi Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-1-10 2018.11
Quality improvement of Super-Resolution Image for Video Coding

S. Umeda, N. Yano, H. Watanabe

2018 Picture Coding Symposium, 2018 Image Media Processing Symposium (PCSJ/IMPS2018) P-1-8 2018.11
Sentiment Analysis on Text of Comics

S. Sheng, H. Watanabe

IEICE Society Conference A-10-18 2018.09
The Performance of Self-Enhancement Method in Optical Flow-Based Motion Estimation for Sequential Fisheye Images with Sudden Movement of the Objects

Arief Suryadi Satyawan, Junichi Hara, Hiroshi Watanabe

IEICE Society Conference BS-7-8 2018.09
A Study on Improvement of Comic Character Facial Image Clustering System Using CNN

H. Yanagisawa, T. Yamashita, H. Watanabe

ITE Annual Conference 12B-2 2018.08
Scene Flow Estimation from Squential Stereo Fisheye Images

Arief Suryadi Satyawan, Junichi Hara, Hiroshi Watanabe

International Conference on Signals and Systems 2018, (ICSygSys2018) DSP2-5 2018.05
Data Driven Cyber-Physical System for Landslide Detection

Zhi Liu, Toshitaka Tsuda, Hiroshi Watanabe, Satoko Ryuo, Nagateru Iwasawa

Mobile Networks and Applications No.11036 1 - 12 2018.03

　View Summary

Natural disaster is one of the most important research topics worldwide. In this paper, a data driven cyber-physical system is introduced to detect landslides. This system is composed of Wi-Sun acceleration sensors, which can detect the acceleration of the nearby environment in 3D domain, and the sensors are linked with the router (act as ’sink’ node) via Wi-Sun transmission (i.e. IEEE802.15.4g). The details of the detection system are explained and the landslide detection mechanism with low computational complexity is proposed. A traffic reduction method is proposed thereafter to help reduce the data needed for transmission by exploring the intra-correlations of the sensor data. This method can save the energy consumption without degrading the detection performance. Field test is conducted and the results show that the landslide can be detected and amount of data to be transmitted can be reduced, which verifies the system’s effectiveness.

DOI

Scopus

40

Citation

(Scopus)
Deformation Influence Evaluation of Character Face Detection Using CNN

T. Yamashita, H. Yanagisawa, H. Watanabe

IPSJ National Convention 5Y-03 2018.03
Position Estimation and Distance Measurement from Omni-directional Cameras

T. Aoki, H. Watanabe, H. Song

IPSJ National Convention 2U-05 2018.03
3D Reconstruction Using LiDAR and Omni-directional Image

H. Akatsuka, H. Watanabe

IPSJ National Convention 2U-04 2018.03
Quantize parameter dependent super resolution characteristics in Future Video Coding

S. Umeda, N. Yano, H. Watanabe, T. Ikai, T. Chujo, N. Ito

IPSJ National Convention 2U-03 2018.03
Quality Improvement of FVC Encoded Image Using Super-Resolution Considering Feature Directionality

N. Yano, S. Umeda, H. Watanabe, T. Ikai, T. Chujo, N. Ito

IPSJ National Convention 2U-02 ( 1 ) 259 - 260 2018.03

CiNii
Coincidence Evaluation of Multiple People’s Dance Using OpenPose

K. Inada, T. Ishikawa, H. Watanabe

IPSJ National Convention 2U-01 2018.03
Batting Statistics Prediction by Analyzing Swing of Baseball Players

S. Yokoi, T. Ishikawa, H. Watanabe

IEICE General Conference D-12-45 2018.03
Improvement of Walking Detection Method for Texting While Walking Detection Using Pose Estimation

K. Kato, H. Watanabe

IEICE General Conference D-12-44 2018.03
Prediction of Next Day’s Sea-Level Pressure by Support Vector Regression

Tin Nilar Lin, Hiroshi Watanabe

IEICE General Conference D-20-11 2018.03
Instance Segmentation on Omni-directional Images Based on Mask R-CNN

Mengcheng Song, Junichi Hara, Hiroshi Watanabe

IEICE General Conference BS-2-13 2018.03
An Improvement of Motion Estimation on Sequential Fisheye Images

Arief Suryadi Satyawan, Junichi Hara, Hiroshi Watanabe

IEICE General Conference BS-2-12 2018.03

J-GLOBAL
Robust 3D Reconstruction with Omni-directional Camera based on Structure from Motion

Mengcheng Song, Junich Hara, Hiroshi Watanabe

International Workshop on Advanced Image Technology (IWAIT2018) No.105, pp.1-4 2018.01
Deep Learning Based Super-Resolution for Future Video Coding

Seiya Umeda, Niai Yano, Hiroshi Watanabe, Tomohiro Ikai, Takeshi Chujoh, Norio Ito

International Workshop on Advanced Image Technology (IWAIT2018) No.89, pp.1-4 2018.01
Scene Flow from Stereo Fisheye Images

Arief Suryadi Satyawan, Junichi Hara, Hiroshi Watanabe

International Workshop on Advanced Image Technology (IWAIT2018) No.77, pp.1-4 2018.01
A Study on Object Detection Method from Manga Images Using CNN

Hideaki Yanagisawa, Takuro Yamashita, Hiroshi Watanabe

International Workshop on Advanced Image Technology (IWAIT2018) No.16, pp.1-4 2018.01
A Study on Texting While Walking Recognition Method Using Pose Estimation and Grasped Object Recognition

K. Kato, H. Watanabe

22B ( 2 ) 2017.12
A Study on Bike Racer Identification Using Joint Position Information

25C ( 5 ) 2017.12
A Study on Improvement of Manga Character Face Image Clustering

22B ( 8 ) 2017.12
Super-Resolution Technology Considering Directionality of Image

YANO Niai, UMEDA Seiya, WATANABE Hiroshi, IKAI Tomohiro, CHUJOH Takeshi, ITO Norio

PROCEEDINGS OF THE ITE WINTER ANNUAL CONVENTION 14B ( 2 ) 14B-2 2017.12

　View Summary

We show that in A+, ScSR and SRCNN, which are the state-of-the-art Super-Resolution technologies, the level of image quality improvement varies according to the “directionality” within the patch of the input image. Based on this observation, we propose a method to further improve quality of the image employing the “directionality”.

DOI
Weather data estimation by sensitive features selection

Tin Nilar Lin, H. Watanabe

13B ( 6 ) 2017.12
多段ニューラルネットワークを用いた動物の品種識別

渡部, 渡辺

2017年映像情報メディア学会冬季大会 13B ( 1 ) 2017.12
Comparison of Manga Object Detection Method Using CNN

H. Yanagisawa, H. Watanabe

P1 ( 10 ) 2017.11
Deep Learning Face Parts Game

P5 ( 2 ) 2017.11
姿勢情報を用いた自転車選手識別手法に関する一検討

高木, 石川,渡辺

映像情報メディア学会メディア工学研究会（スポーツ情報処理時限研究会合同開催） ME2017 ( 121 ) 25 - 27 2017.11
畳込みニューラルネットによるマンガオブジェクト認識メカニズムの一検討

柳澤, 山下,渡辺

電子情報通信学会，パターン認識・メディア理解研究会 PRMU2017 ( 79 ) 2017.10
V-SLAMとマーカARを併用したペン型デバイスによる空中描画システム

小林, 石川,渡辺

映像情報メディア学会年次大会 34B ( 3 ) 2017.09
ステレオ全天球カメラからの距離計測の基礎検討

青木,松田, 原,渡辺

映像情報メディア学会年次大会 33B ( 5 ) 2017.09

J-GLOBAL
姿勢情報を用いた自転車選手識別の基礎検討

高木, 石川,渡辺

映像情報メディア学会年次大会 14D ( 4 ) 2017.09
関節角度情報を用いた歩きスマホ認識

加藤,渡辺

映像情報メディア学会年次大会映像情報メディア学会年次大会 13D ( 4 ) 2017.09
ボロノイ図に基づく3次元優勢領域によるパスコース評価について

横井,浅見, 石川,渡辺

映像情報メディア学会年次大会 13D ( 3 ) 2017.09
Ship Classification Using Faster Region Convolution Neural Network (Faster R-CNN) for Automatic Identification of Marine Vessels

K. J. Ngeno, H. Yanagisawa, H. Watanabe

H ( 39 ) 2017.09
Data Augmentation Using DCGAN Considering Generated Image Quality

H ( 38 ) 2017.09
Walking Detection for Texting While Walking Recognition by Pose Estimation

H ( 36 ) 2017.09
Water Level Prediction for Disaster Management Using Machine Learning Models

Tin Nilar Lin, H. Watanabe

H ( 29 ) 2017.09
Motion Estimation on Fish-Eye Images Using Modified Motion Model

Arief Suryadi Satyawan, J. Hara, H. Watanabe

H ( 23 ) 2017.09
Deep Learning特徴量を用いたマンガキャラクター顔画像の分類

柳澤,渡辺

FIT2017(第16回科学技術フォーラム) H ( 1 ) 2017.09
姿勢推定による肘と肩の角度情報を用いた歩きスマホ認識

加藤, 渡辺

映像情報メディア学会スポーツ情報処理時限研究会 4 ( 7 ) 2017.07
3次元優勢領域図を利用したパスコースの決定

横井, 浅見, 石川, 渡辺

映像情報メディア学会スポーツ情報処理時限研究会 4 ( 6 ) 2017.07
[Foreword] Welcome to the Special Section on Sports-Information Technology

Watanabe Hiroshi

ITE Transactions on Media Technology and Applications 5 ( 2 ) 35 - 35 2017.04

CiNii
Trajectory Visualization of Ego-Motion Videos with Pedestrian Based on Monocular Visual Odometry and Machine Learning

Y. Zhang, H. Watanabe

IEICE General Conference BS-1 ( 8 ) 2017.03
Motion Estimation on Fish-Eye Images

A. S. Satyawan, J. Hara, H. Watanabe

IEICE General Conference BS-1 ( 9 ) 2017.03
Ship Identification for Port Security using Faster Region Convolution Neural Network

K. J. Ngeno, Hiroshi Watanabe

IEICE General Conference BS-1 ( 12 ) 2017.03
X-means法を用いたマンガキャラクターの自動分類に関する検討

柳澤, 渡辺

IEICE総合大会 D-12 ( 40 ) 2017.03
Recognition of Panel Structure in Comic Images Using Faster R-CNN

H. Yahagisawa, H. Watanabe

International Workshop on Image Electronics and Visual Computing 2017 4C ( 2 ) 2017.03
Data Augmentation Using DCGAN for Breed Identification

H. Watabe, H. Watanabe

International Workshop on Image Electronics and Visual Computing 2017 4C ( 3 ) 2017.03 [Refereed]
Phase Competition and Superconductivity in κ-(BEDT-TTF)₂X: Importance of Intermolecular Coulomb Interactions

Watanabe Hiroshi, Seo Hitoshi, Yunoki Seiji

Journal of the Physical Society of Japan 86 ( 3 ) 2017.02

CiNii
ボロノイ図を用いたサッカーの3次元優勢領域の決定

横井, 石川, 渡辺

映像情報メディア学会スポーツ情報処理時限研究会 3 ( 1 ) 2017.02
Welcome to the special section on sports-information technology

Hiroshi Watanabe

ITE Transactions on Media Technology and Applications 5 ( 2 ) 35 2017

DOI

Scopus
Adaptive Video Streaming in Hybrid Landslide Detection System with D-S Theory

Zhi Liu, Kenji Kanai, Masaru Takeuchi, Toshitaka Tsuda, Hiroshi Watanabe

2017 IEEE INTERNATIONAL CONFERENCE ON SMART COMPUTING (SMARTCOMP) 352 - 357 2017 [Refereed]

　View Summary

Disaster detection is an important research topic and draws great attentions from both industry and academia. In this paper, we study the hybrid landslide detection system, which utilizes the video surveillance camera and multiple kinds of sensors and can detect the landslide automatically. Edge processing is adopted in this hybrid system to fuse the sensor data based on the Dempster-Shafer (D-S) theory, i.e. utilizing multiple sensors' information to calculate the possibility of the landslide. Edge processing can make faster decisions than the control center, the results of the edge processing are then used to schedule the sensor's transmission frequency and video transmission under the network constraints in this system. The simulation results show that the proposed scheme outperforms the competing schemes in typical network scenarios.
Adaptive Video Streaming in Hybrid Landslide Detection System with D-S Theory

Zhi Liu, Kenji Kanai, Masaru Takeuchi, Toshitaka Tsuda, Hiroshi Watanabe

2017 IEEE INTERNATIONAL CONFERENCE ON SMART COMPUTING (SMARTCOMP) 352 - 357 2017 [Refereed]

　View Summary

Disaster detection is an important research topic and draws great attentions from both industry and academia. In this paper, we study the hybrid landslide detection system, which utilizes the video surveillance camera and multiple kinds of sensors and can detect the landslide automatically. Edge processing is adopted in this hybrid system to fuse the sensor data based on the Dempster-Shafer (D-S) theory, i.e. utilizing multiple sensors' information to calculate the possibility of the landslide. Edge processing can make faster decisions than the control center, the results of the edge processing are then used to schedule the sensor's transmission frequency and video transmission under the network constraints in this system. The simulation results show that the proposed scheme outperforms the competing schemes in typical network scenarios.

DOI

Scopus

2

Citation

(Scopus)
Faster R-CNNを用いたマンガ画像の構造解析

柳澤, 渡辺

P-2 ( 10 ) 2016.11
線路脇カメラと鉄道車載カメラによる障害物検出の検討

加藤, 渡辺

P-3 ( 16 ) 2016.11
レンズ歪補正動き補償方式に関する一検討

高木, 渡辺, 河村, 内藤

P-3 ( 19 ) 2016.11
DCGANによる生成画像品質の猫品種識別性能に及ぼす影響について

渡部, 渡辺

P-4 ( 16 ) 2016.11
DCGANを用いたデータオーギュメンテーションによる猫の品種識別について

渡部, 渡辺

2016映像情報メディア学会年次大会 24D ( 5 ) 2016.09
レンズ歪ブロックを用いた動き補償方式の検討

高木, 渡辺, 河村, 内藤

FIT2016(第15回科学技術フォーラム) I ( 033 ) 2016.09
Faster R-CNNを用いたマンガ画像からのメタデータ抽出

柳澤, 渡辺

2016映像情報メディア学会年次大会 14B ( 1 ) 2016.09
Trajectory data visualization of sports video based on SLAM

Y. Zhang, H. Watanabe

ITE Annual Convention 2016 34C ( 1 ) 2016.09
BS-3-27 Empty Scene and Non-empty Scene Detection from Wi-Sun Acceleration Sensor(BS-3. Advanced Networking Technologies for Innovative Information Networks)

Liu Zhi, Sigg Stephan, Tsuda Toshitaka, Watanabe Hiroshi

Proceedings of the IEICE General Conference 2016 ( 2 ) "S - 64"-"S-65" 2016.03

　View Summary

Wireless sensor networks (WSNs) are widely deployed to help detect disaster such as landslides, etc. in countries like Japan due to their efficiency and low cost. But there are various problems in these systems, such as the different kind of noise (such as random visit from animals, which leads to unnecessary false alarms). Orthogonally, context aware activity recognition draws increasing attention recently, where different environmental or martial sensors are adopted for activity recognition in various application scenarios. This paper studies a disaster detection system utilizing the Wi-Sun acceleration sensors together with the Wi-Sun signal fluctuation and considers the detection of human beings in the scene of interest. Hence, the false alarms introduced by the animals' walk in, etc. could be detected using similar method. During the detection process, not only the wireless signal but also the acceleration value collected by the sensors are adopted for a better detection result. Then feature values are calculated for event detection based on the data samples. K nearest neighbors (KNN) is used to classify the events in the scene of interest: empty and non-empty. The detection results are promising as shown in the result section and the proposed method is applicable to a real landslide to avoid animal- induced false alarms.

CiNii
D-21-4 A study on creation method for e-comic trailers

NOGUCHI Taki, WATANABE Hiroshi

Proceedings of the IEICE General Conference 2016 ( 2 ) 217 - 217 2016.03

CiNii
D-21-3 A study on major character identification in comics

NAGAO Kazuki, WATANABE Hiroshi

Proceedings of the IEICE General Conference 2016 ( 2 ) 216 - 216 2016.03

CiNii
D-12-12 A Study of Multi-view Face Detection for Characters in Comic Images

Yanagisawa Hideaki, Watanabe Hiroshi

Proceedings of the IEICE General Conference 2016 ( 2 ) 81 - 81 2016.03

CiNii
D-11-53 A Study on Motion Compensation Method with Radial Distortion Correction

Takagi Masanori, Watanabe Hiroshi, Kawamura Kei, Naito Sei

Proceedings of the IEICE General Conference 2016 ( 2 ) 53 - 53 2016.03

CiNii
D-11-20 Utilization of Background Subtraction Method for Railway Operation Image

KATO Kimimaru, WATANABE Hiroshi

Proceedings of the IEICE General Conference 2016 ( 2 ) 20 - 20 2016.03

CiNii
Empty Scene and Non-empty Scene Detection from Wi-Sun Acceleration Sensor

Z. Liu, S. Sigg, T. Tsuda, H. Watanabe

IEICE General Conference BS-3 ( 27 ) 2016.03
鉄道運行映像に対する背景差分法の適用

加藤, 渡辺

電子情報通信学会総合大会 D-11 ( 20 ) 2016.03
レンズ歪み補正付き動き補償方式の検討

高木, 渡辺, 河村, 内藤

D-11 ( 53 ) 2016.03
マンガキャラクターのマルチビュー顔検出に関する検討

柳澤, 渡辺

D-11 ( 12 ) 2016.03
コミックにおける主要キャラクター同定の検討

長尾, 渡辺

D-21 ( 3 ) 2016.03
電子コミックトレーラーの作成手法の検討

野口, 渡辺

電子情報通信学会総合大会 D-21 ( 4 ) 2016.03
Convolutional Pose Machines

Shih-En Wei, Varun Ramakrishna, Takeo Kanade, Yaser Sheikh

2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR) D-12 ( 56 ) 4724 - 4732 2016 [Refereed]

　View Summary

Pose Machines provide a sequential prediction framework for learning rich implicit spatial models. In this work we show a systematic design for how convolutional networks can be incorporated into the pose machine framework for learning image features and image-dependent spatial models for the task of pose estimation. The contribution of this paper is to implicitly model long-range dependencies between variables in structured prediction tasks such as articulated pose estimation. We achieve this by designing a sequential architecture composed of convolutional networks that directly operate on belief maps from previous stages, producing increasingly refined estimates for part locations, without the need for explicit graphical model-style inference. Our approach addresses the characteristic difficulty of vanishing gradients during training by providing a natural learning objective function that enforces intermediate supervision, thereby replenishing back-propagated gradients and conditioning the learning procedure. We demonstrate state-of-the-art performance and outperform competing methods on standard benchmarks including the MPII, LSP, and FLIC datasets.

DOI
R-CNNを用いたマンガキャラクター検出に関する一検討

柳澤, 渡辺

I-4 ( 12 ) 1 - 2 2015.11
Cat and dog face detection using deformable part model

H. Watanabe, H. Zhang

PCSJ-IMPS Symposium I-3 ( 12 ) 1 - 2 2015.11
マンガ要約手法に関する一考察

藤岡, 渡辺

電子情報通信学会ソサイティ大会 A20 ( 11 ) 2015.09
A-20-11 A Study on Digesting Method for Comics

FUJIOKA Kyohei, WATANABE Hiroshi

Proceedings of the IEICE Engineering Sciences Society/NOLTA Society Conference 2015 151 - 151 2015.08

CiNii
Sensor Data Characteristics in the Wireless Train Track Unusual Notification Sensor Network

Z. Liu, T. Tsuda, H. Watanabe

IEICE Technical Report on Communication Quality CQ2015 ( 39 ) 101 - 106 2015.07

CiNii
An effective normalization algorithm at deep learning for passenger detection in railway operation videos

J. Zhang, H. Watanabe

The 1st International Conference on Advanced Imaging ( T109-2 ) 2015.06 [Refereed]
多層畳み込みニューラルネットワークを用いた人物認識における正規化効果

張, 石井, 渡辺

電子情報通信学会総合大会 D-12-24 2015.03
マンガの複数キャラクターに対する顔検出率について

柳澤, 石井, 渡辺

電子情報通信学会総合大会 D-12-31 2015.03
ディープラーニングを用いたオブジェクト認識における整流・マックスプーリング効果

張, 石井, 渡辺

情報処理学会全国大会 2P-08 2015.03
On face detection rate for characters in comic images

Yanagisawa Hideaki, Ishii Daisuke, Watanabe Hiroshi

Proceedings of the IEICE General Conference D-12 ( 31 ) 83 - 83 2015.02

CiNii
Effect of Normalization at Deep Convolutional Neural Networks for Human Recognition

Zhang Jie, Ishii Daisuke, Watanabe Hiroshi

Proceedings of the IEICE General Conference D-12 ( 24 ) 76 - 76 2015.02

CiNii
Traffic Deduction Exploring Sensor Data's Intra-Correlations in Train Track Monitoring WSN

Zhi Liu, Toshitaka Tsuda, Hiroshi Watanabe

2015 IEEE SENSORS 7 ( 171 ) 1712 - 1715 2015 [Refereed]

　View Summary

WSNs are good options to help monitor the scene of interest and notify the unusual happening to control center. But sensors' high sampling rates lead to tremendous network traffic over the bandwidth-limited and energy-critical WSNs, hence how to reduce the network traffic while maintaining the unusual events monitoring function becomes important. In this paper, we investigate the intra-correlations of the data generated by each sensor at different time instances. And we propose a traffic deduction algorithm exploring the sensor data's intra-correlations which could reduce the data volume significantly and guarantee the parameters needed for unusual detection are delivered.
Traffic Deduction Exploring Sensor Data's Intra-Correlations in Train Track Monitoring WSN

Zhi Liu, Toshitaka Tsuda, Hiroshi Watanabe

2015 IEEE SENSORS 1712 - 1715 2015 [Refereed]

　View Summary

WSNs are good options to help monitor the scene of interest and notify the unusual happening to control center. But sensors' high sampling rates lead to tremendous network traffic over the bandwidth-limited and energy-critical WSNs, hence how to reduce the network traffic while maintaining the unusual events monitoring function becomes important. In this paper, we investigate the intra-correlations of the data generated by each sensor at different time instances. And we propose a traffic deduction algorithm exploring the sensor data's intra-correlations which could reduce the data volume significantly and guarantee the parameters needed for unusual detection are delivered.

DOI

Scopus

7

Citation

(Scopus)
マンガの構成要素に基づく自動シーン分割処理に関する一検討

石井, 柳澤, 三原, 永森, 渡辺

情報処理学会AVM研究会研究報告 Vol.2014-AVM87 ( No.15 ) 1 - 4 2014.12
A basic study on human detection for train operation video

J. Zhang, H. Watanabe

PCSJ-IMPS 2014 I-5-01 2014.11
放送の「こと」づくり

渡辺裕

映像情報メディア学会誌 68 ( 10 ) k18 - k18 2014.10

DOI CiNii
2値線画像解析のための2値方向パターンヒストグラム特徴量HBOP

石井, 渡辺

早稲田大学GITS/GITI紀要2013-1014 33 - 44 2014.10

CiNii
Face detection for comic images with deformable part model

H. Yanagisawa, D. Ishii, H. Watanabe

The 4th International Workshop on Image Electronics and Visual Computing 2014 (IEVC2014) 4A-1 ( 2 ) 95 - 100 2014.10

CiNii
マンガにおけるHOG+AdaBoostによる顔画像検出の性能評価

陳, 柳澤, 張, 石井, 渡辺

映像情報メディア学会年次大会 17-4 17 - 4-1_-_17-4-2_ 2014.09

　View Summary

In recent years, Japanese comics have been very popular in the world. In this paper, we applied the HOG and AdaBoost algorithm to detect character's face in comic images. We have recognized that learning process requires at least 80 times repeatedly and the detection success rate converges in about 65%.

DOI CiNii
マンガ画像からの顔検出におけるパーツ特徴量の一検討

柳澤, 石井, 陳, 渡辺

映像情報メディア学会年次大会 17-9 17 - 9-1_-_17-9-2_ 2014.09

　View Summary

In recent years, studies of extracting meta-data from comic images has been focused on for query and search application of e-comics. In this paper, we propose to apply Deformable Part Model, which is a detection method using parts placement of an object, to detect character' faces in comic images.

DOI CiNii
A Note on Face Detection of Comic Image with Different Background

M. Chen, H. Yanagisawa, D. Ishii, H. Watanabe

2014 IEICE General Conference BS-1-47 1 - 2 2014.03
A note on content caching and delivery based on AVC and SVC

M. Takafuji, H. Nakazato, H. Watanabe

情報処理学会AVM研究会研究報告 Vol.2014-AVM84 ( No.3 ) 1 - 5 2014.02

　View Summary

Content cache effectiveness in video distribution is mathematically analyzed. Targeted video distribution models are two types. One is simulcast, the other is scalable approach. Simple two layer model is used for the analysis. Caching effect is studied focusing on the amount of required transmission data based on some assumptions with a certain cache hit probability. The result shows that the scalable approach cannot improve the caching effect.

CiNii
2値線画特徴量を用いたマンガ画像解析に関する一検討

石井, 張, 石上, 渡辺

情報処理学会AVM研究会研究報告 Vol.2014-AVM84 ( No.5 ) 1 - 4 2014.02

　View Summary

Comic image was drawn into complex pattern using binary line. In conventional research, HOG it can treat image edge was utilized for to analyze Comic image as image feature. On the other hand HOG feature has redundant dimension for to treat binary line image. Moreover, diagonal line on binary image was consisted by combination of vertical and horizontal edge. Thus it is hard to get line angle. Histograms of Binary Oriented Pattern(HBOP) was proposed for to analyze binary line image by us. In this research, We analyze comic image using HBOP. Then, comparison results between HOG with HBOP on Comic image analysis was derived by examination.

CiNii
“HTML5 CanvasおよびSVGにおける自動選択アルゴリズムを用いた描画パフォーマンスの最適化

山本, 渡辺

情報処理学会AVM研究会研究報告 Vol.2014-AVM84 ( No.6 ) 1 - 4 2014.02

　View Summary

In HTML5, Canvas dealing with data of the bitmap and Inline SVG dealing with Data for vector are supported. Thus we can directly describe image or graphic in HTML, by the JavaScript. However, drawing performance is not good compared with the plugins such as Flash or Java. In this paper, we sinvestigate the performance of traditional approaches to increase drawing speed. Finally, we propose an algorithm to select the best approach automatically.

CiNii
マンガからの自動人物検出と識別に関する一検討

石井大祐, 渡辺裕

画像電子学会論文誌 Vol. 42 ( No. 4 ) 457 - 465 2013.07

　View Summary

Recently, comics has been contents with many user not only Japan but also foreign countries. Moreover, demand for digital comic is increased by digital book equipment developing and growth of e-book market. Comics have very complex construction because a lot of elements that picture, texts, onomatopoeia, and so on are drawn on a plane. To extract elements and characters as metadata, we can increase usability of digitized comics. Comics generally have been drawn white and black, so the feature of image is different with natural picture. Thus, to analyze comic character, we must select method that is able to treat comic image feature. In this paper, we propose specified character detection methods on a comic page by eye detection, face detection, and person recognition that utilize HOG (Histograms of Oriented Gradient) and SVM (Support Vector Machine). Finally, we obtained 0.93 maximum precision for specified character detection by proposed method.

DOI CiNii
マンガ固有の特徴を利用したマンガ登場人物識別に関する一検討

石井, 山﨑, 渡辺

情報処理学会AVM研究会研究報告 Vol.2013-AVM80 ( No.1 ) 1 - 4 2013.04
マンガ上のキャラクター識別に関する一検討

石井, 山﨑, 渡辺

情報処理学会全国大会 3D-2 2013.03
フーリエ記述子と色情報を用いた野菜識別手法

山﨑, 石井, 渡辺

情報処理学会AVM研究会研究報告 Vol.2012-AVM79 ( No.27 ) 1 - 4 2012.12
Multi Size Eye Detection on Digitized Comic Image

D. Ishii, T. Yamazaki, H. Watanabe

IIEEJ 3rd Image Electronics and Visual Computing Workshop (IEVC 2012) 2012.11
マンガ画像における瞳検出の解像度依存性に関する検討

石井, 渡辺

2012年度画像電子学会年次大会(Media Computing Conference 2012) R2-4 2012.06
マンガ画像からの自動キャラクター位置検出に関する検討

石井, 渡辺

Vol.2012-AVM76 ( No.2 ) 1 - 5 2012.02
Interpolation-Free Fractional Pixel Motion Estimation Based on Data Trend Approximation

Chang-Uk Jeong, Hiroshi Watanabe

2012 INTERNATIONAL CONFERENCE ON SIGNALS AND ELECTRONIC SYSTEMS (ICSES) 2012 [Refereed]

　View Summary

High computational complexity of a fractional pixel motion estimation (FME) module can in no way be negligible although the module improves visual quality after the integer pixel motion estimation (IME) process. Most of the conventional FME methods must include the interpolation procedure to form fractional pixel search points from the information of the integer pixel matching error costs. The interpolation, however, requires a considerable memory usage and certain amount of processing time. In this paper, therefore, we propose interpolation-free fractional pixel motion estimation techniques by using the data trend approximation and the error cost scaling. The results of our simulation show that the proposed methods produce similar or better performances compared with the existing FME methods, whereas they do not need any additional search points.
A Modified Parabolic Prediction Based Fractional Pixel Motion Estimation Using Preset Corrector

Chang-Uk Jeong, Hiroshi Watanabe

18TH ASIA-PACIFIC CONFERENCE ON COMMUNICATIONS (APCC 2012): GREEN AND SMART COMMUNICATIONS FOR IT INNOVATION 526 - 527 2012 [Refereed]

　View Summary

In general, a motion estimator in today's video coding includes fractional pixel motion estimation (FME) as well as integer pixel motion estimation (IME). FME can get better quality performance at the cost of higher computational complexity than IME alone. In this paper, a modified parabolic prediction based FME for H.264 video coding is proposed. The proposed method uses specific correction coefficients to improve the PSNR performance of the existing prediction based algorithm. In the simulation results, compared with the original algorithm, our technique shows more accurate predictive power, meanwhile, it does not require the interpolation process.
Interpolation-Free Fractional Pixel Motion Estimation Based on Data Trend Approximation

Chang-Uk Jeong, Hiroshi Watanabe

2012 INTERNATIONAL CONFERENCE ON SIGNALS AND ELECTRONIC SYSTEMS (ICSES) 52 - 62 2012 [Refereed]

　View Summary

High computational complexity of a fractional pixel motion estimation (FME) module can in no way be negligible although the module improves visual quality after the integer pixel motion estimation (IME) process. Most of the conventional FME methods must include the interpolation procedure to form fractional pixel search points from the information of the integer pixel matching error costs. The interpolation, however, requires a considerable memory usage and certain amount of processing time. In this paper, therefore, we propose interpolation-free fractional pixel motion estimation techniques by using the data trend approximation and the error cost scaling. The results of our simulation show that the proposed methods produce similar or better performances compared with the existing FME methods, whereas they do not need any additional search points.
Automatic Scale Detection for Contour Fragment Based on Difference of Curvature

Kei Kawamura, Daisuke Ishii, Hiroshi Watanabe

IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS E94D ( 10 ) 1998 - 2005 2011.10 [Refereed]

　View Summary

Scale-invariant features are widely used for image retrieval and shape classification. The curvature of a planar curve is a fundamental feature and it is geometrically invariant with respect it the coordinate system. The curvature-based feature varies in position when multiscale analysis is performed. Therefore, it is important to recognize the scale in order to detect the feature point. Numerous shape descriptors based on contour shapes have been developed in the field of pattern recognition and computer vision. A curvature scale-space (CSS) representation cannot be applied to a contour fragment and requires the tracking of feature points. In a gradient-based curvature computation, although the gradient computation considers the scale, the curvature is normalized with respect to not the scale but the contour length. The scale-invariant feature transform algorithm that detects feature points from an image solves similar problems by using the difference of Gaussian (DoG). It is difficult to apply the SIFT algorithm to a planar curve for feature extraction. In this paper, an automatic scale detection method for a contour fragment is proposed. The proposed method detects the appropriate scales and their positions on the basis of the difference of curvature (DoC) without the tracking of feature points. To calculate the differences, scale-normalized curvature is introduced. An advantage of the DoC algorithm is that the appropriate scale can be obtained from a contour fragment as a local feature. It then extends the application area. The validity of the proposed method is confirmed by experiments. The proposed method provides the most stable and robust scales of feature points among conventional methods such as curvature scale-space and gradient-based curvature.

DOI

Scopus
Sub-pixel position prediction techniques based on Bazier curve

C. Jeong, H. Watanabe

IPSJ AVM Technigal Report Vol.2011-AVM73 ( No.8 ) 1 - 6 2011.07

CiNii
Automatic preview generation of comic episodes for digitized comic search.

Keiichiro Hoashi, Chihiro Ono, Daisuke Ishii, Hiroshi Watanabe

MM’11 Proceedings of the 19th ACM international conference on Multimedia 1489 - 1492 2011

DOI

Scopus

8

Citation

(Scopus)
A Study on Frame Position Detection of Digitized Comics Images

D. Ishii, H. Watanabe

Workshop Picture Coding Symposium WP3-17 124 - 125 2010.12
A Linear Model-Based Sub-Pixel Motion Estimation Method For H.264/AVC Standard

C. Jeong, H. Watanabe

Workshop Picture Coding Symposium WP1-5 11 - 12 2010.12
多重記述符号化における冗長係数群の最適化手法

石川, 渡辺

情報処理学会AVM 研究会研究報告 Vol.2010-AVM71 ( No.21 ) 1 - 6 2010.12
A Vertically Symmetrical Linear Model-Based Fractional-Pixel Motion Estimation Algorithm for H.264/AVC Encoder

C. Jeong, H. Watanabe

IPSJ SIG Technical Report Vol.2010-AVM71 ( No.20 ) 1 - 5 2010.12
全変動最小化の高速計算手法

河村, 石井, 渡辺

電子情報通信学会論文誌 Vol. J93-D ( No.3 ) 326 - 335 2010.03

CiNii
コミック画像の文字位置特定に関する検討

河村, 石井, 渡辺

電子情報通信学会総合大会 D-12-52 2010.03
過完備変換係数の最適化における画質評価に関する考察

石川, 渡辺, 坂東, 高村, 如澤

電子情報通信学会総合大会 D-11-87 2010.03
四元数を用いた画像符号化方式の検討

石川, 渡辺

画像符号化シンポジウム PCSJ2009 P-4.03 75 - 76 2009.10
領域分割に基づくベクタ表現を用いた画像符号化方式の検討

河村, 石井, 渡辺

画像符号化シンポジウムPCSJ2009 P-3.09 2009.10
フレーム理論に基づく信号の冗長度制御方式

石川, 渡辺

FIT2009 I-045 331 - 332 2009.09
線画の混在する画像におけるテキスト領域抽出の改善手法

河村, 石井, 渡辺

FIT2009 H-027 2009.09
[招待講演]モバイルディジタルテレビジョンのためのビデオ符号化

渡辺

信学技報 MoMuC2009-28 vol.109 ( no.204 ) 9 - 14 2009.09
フレーム展開による画像の多重記述符号化に関する検討

石川, 渡辺

情報処理学会 AVM 研究会研究報告 Vol.2009-AVM65 ( No.16 ) 1 - 6 2009.07
コミック画像のコマ分割処理における制御パラメータに関する検討

石井, 河村, 渡辺

電子情報通信学会パターン認識・メディア理解研究会 PRMU2009-34 ( 34 ) 187 - 192 2009.05

CiNii
[チュートリアル招待講演]MPEGビデオ符号化方式

渡辺裕

電子情報通信学会技術研究報告 NS2009-14 71 - 76 2009.04

CiNii
グラデーションメッシュ最適化手法の検討

河村, 石井, 渡辺

電子情報通信学会総合大会 D-11-36 2009.03
角度と複比を用いた複数車線検出に関する一検討

谷, 河村, 渡辺

電子情報通信学会総合大会 D-11-68 2009.03
並列木複素ウェーブレット変換のブロック歪に関する考察

石川, 渡辺, 坂東, 高村, 上倉, 八島

電子情報通信学会総合大会 AS-3-3 2009.03
Noise Shapingを用いた並列木複素wavelet変換符号化に関する基礎的考察

坂東, 高村, 上倉, 八島, 石川, 渡辺

電子情報通信学会総合大会 D-11-57 2009.03
高ダイナミックレンジ画像生成におけるゴースト除去手法に関する検討

高木, 渡辺

電子情報通信学会総合大会 D-11-51 2009.03
Automatic Scale Detection Based on Difference of Curvature

K. Kawamura, D. Ishii, H. Watanabe

2009 International Workshop on Advanced Image Technology (IWAIT 2009) 2009.01

CiNii
グラデーションメッシュによる画像符号化の基礎検討

河村, 石井, 渡辺

画像符号化シンポジウム PCSJ2008 P-5.04 2008.10
アンチエイリアシングを考慮したベクタ表現による濃淡画像符号化に関する一検討

石井, 河村, 渡辺

画像符号化シンポジウム PCSJ2008 P-2.04 2008.10
画像の不確定性符号化におけるレート制御方式に関する検討

石川, 渡辺

画像符号化シンポジウム PCSJ2008 P-2.02 2008.10

CiNii
フレーム展開による画像の不確定性符号化に関する検討

石川, 渡辺

FIT2008 I-016 2008.09
複比を用いた複数車線検出に関する検討

谷, 河村, 渡辺

FIT2008 H-040 2008.09
スケール不変な曲線特徴に関する検討

河村, 石井, 渡辺

河村, 石井, 渡辺 H-025 2008.09
移動物体が存在する複数画像からの高ダイナミックレンジ画像生成に関する検討

高木, 渡辺

情報処理学会AVM研究会 2008AVM62-03 2008.09
SPORT: An algorithm for Divisible Load Scheduling with result collection on heterogeneous systems

Abhay Ghatpande, Hidenori Nakazato, Olivier Beaumont, Hiroshi Watanabe

IEICE TRANSACTIONS ON COMMUNICATIONS E91B ( 8 ) 2571 - 2588 2008.08

　View Summary

Divisible Load Theory (DLT) is an established mathematical framework to study Divisible Load Scheduling (DLS). However, traditional DLT does not address the scheduling of results back to source (i.e., result collection), nor does it comprehensively deal with system heterogeneity. In this paper, the DLSRCHETS (DLS with Result Collection on HET-erogeneous Systems) problem is addressed. The few papers to date that have dealt with DLSRCHETS, proposed Simplistic LIFO (Last In, First Out) and FIFO (First In, First Out) type of schedules as solutions to DLSRCHETS. In this paper, a new polynomial time heuristic algorithm, SPORT (System Parameters based Optimized Result Transfer), is proposed as a solution to the DLSRCHETS problem. With the help of simulations, it is proved that the performance of spoRT is significantly better than existing algorithms. The other major contributions of this paper include, for the first time ever, (a) the derivation of the condition to identify the presence of idle time in a FIFO schedule for two processors, (b) the identification of the limiting condition for the optimality Of FIFO and LIFO schedules for two processors, and (c) the introduction of the concept of equivalent processor in DLS for heterogeneous systems with result collection.

DOI

Scopus

6

Citation

(Scopus)
フレーム展開による信号の冗長度制御方式に関する検討

石川, 渡辺

電子情報通信学会画像工学研究会技術報告 IE2008-43 Vol.108 ( No.127 ) 61 - 66 2008.07

　View Summary

We propose a new method to control signal redundancy by frame expansion for multiple description coding. A conventional method is to partition coefficients of Discrete Wavelet Transform (DWT) using Lattice in the frequency domain. Redundancy is introduced by distance of the wavelet coefficients based on spatial redundancy in the transform domain of the signal. Proposed method introduces redundancy to wavelet coefficients by frame expansion. Experimental results show that proposed method enhances the error resilience of image transmission under packet loss when compared to the conventional method.

CiNii
Analysis of Divisible Load Scheduling with result collection on heterogeneous systems

Abhay Ghatpande, Hidenori Nakazato, Olivier Beaumont, Hiroshi Watanabe

IEICE TRANSACTIONS ON COMMUNICATIONS E91B ( 7 ) 2234 - 2243 2008.07 [Refereed]

　View Summary

Divisible Load Theory (DLT) is an established framework to study Divisible Load Scheduling (DLS). Traditional DLT ignores the result collection phase, and specifies no solution to the general case where both the network speed and computing capacity of the nodes are heterogeneous. In this paper, the DLS with Result Collection on HETerogeneous Systems (DLSRCHETS) problem is formulated as a linear program and analyzed. The papers to date that have dealt with result collection, proposed Simplistic LIFO (Last In, First Out) and FIFO (First In, First Out) type of schedules as solutions. The main contributions of this paper are: (a) A proof of the Allocation Precedence Condition, which is inconsequential in LIFO or FIFO, but is important in a general schedule. (b) A proof of the Idle Time Theorem, which states that irrespective of whether load is allocated to all available processors, in the optimal solution to the DLSRCHETS problem, at the most one processor that is allocated load has idle time, and that the idle time exists only when the result collection begins immediately after the completion of load distribution.

DOI

Scopus

4

Citation

(Scopus)
曲率空間尺度を用いたベクタ変換の評価手法に関する検討

河村, 石井, 渡辺

情報処理学会AVM研究会 2007AVM60-05 2008.03
直感的な曲線操作を実現するベクタ変換の検討

河村, 石井, 渡辺

情報処理学会全国大会 4B-6 2008.03
複素ウェーブレットの疎表現における係数選択手法の検討

高橋, 渡辺

情報処理学会全国大会 3Y-2 2008.03
ベクタ表現を用いた画像のエッジ再現手法に関する一検討

石井, 河村, 渡辺

情報処理学会全国大会 3Y-1 2008.03
グローバル動き補償を用いた背景フレーム補間手法に関する検討

東, 渡辺

情報処理学会全国大会 3W-4 2008.03
全変動ノルムの離散化手法に関する検討

河村, 石井, 渡辺

電子情報通信学会総合大会 D-11-82 2008.03
Divisible Load Scheduling with result collection on heterogeneous systems

Abhay Ghatpande, Hidenori Nakazato, Hiroshi Watanabe, Olivier Beaumont

2008 IEEE INTERNATIONAL SYMPOSIUM ON PARALLEL & DISTRIBUTED PROCESSING, VOLS 1-8 372 - + 2008 [Refereed]

　View Summary

Divisible Load Theory (DLT) is an established mathematical framework to study Divisible Load Scheduling (DLS). However traditional DLT does not comprehensively deal with the scheduling of results back to source (i.e., result collection) on heterogeneous systems. In this paper, the DLSRCHETS (DLS with Result Collection on HETerogeneous Systems) problem is addressed. The few papers to date that have dealt with DLSRCHETS, proposed simplistic LIFO (Last In, First Out) and FIFO (First In, First Out) type of schedules as solutions to DLSRCHETS. In this paper, a new heuristic algorithm, ITERLP, is proposed as a solution to the DLSRCHETS problem. With the help of simulations, it is proved that the performance of ITERLP is significantly better than existing algorithms.
位置ずれを持つ画像に対する高ダイナミックレンジ画像の生成手法に関する検討

高木, 渡辺

情報処理学会AVM研究会 2007AVM59-17 ( 64 ) 101 - 104 2007.12

CiNii
疑似TVノルムの数値計算とその領域分割への応用に関する検討

河村, 石井, 渡辺

情報処理学会AVM研究会 2007AVM59-10 IE2007 - 122 2007.12

CiNii
方向性フィルタバンクによる動画像符号化方式に関する基礎検討

石川, 渡辺

画像符号化シンポジウム PCSJ2007 P-3.02 2007.11
人工的な多値画像のベクター変換

河村, 石井, 渡辺

画像符号化シンポジウム PCSJ2007 P-2.21 2007.11
動画像の時空間ベクタ表現に関する一検討

石井, 河村, 渡辺

画像符号化シンポジウム PCSJ2007 P-1.03 2007.11

CiNii
ベクター表現に適した小数画素精度の改良境界線検出に関する検討

河村, 石井, 渡辺

FIT2007 I-045 2007.09

CiNii
人工的な多値画像のベクタ変換を考慮した領域分割に関する検討

河村, 石井, 渡辺

情報処理学会AVM研究会 2007AVM58-07 2007.09
双対木複素ウェーブレットにおける Energy Compaction の検討

高橋, 渡辺

情報処理学会AVM研究会 2007AVM58-06 2007.09
コミックのコマ分割処理に関する一検討

石井, 河村, 渡辺

電子情報通信学会論文誌D Volume J90-D ( No.7 ) 1667 - 1770 2007.07

CiNii
ベクター表現に適した少数画素精度の改良境界線検出に関する検討

河村, 石井, 渡辺

情報処理学会AVM研究会 2007 ( 22 ) 61 - 64 2007.03

　View Summary

Digitizing paper have been convenient since image input devices are disseminated. Further, an input resolution is lower, an input speed is higher. On the other hand, a resolution conversion is mandatory technique in order to display the image on any panels which have varying resolutions. We have proposed subpixel binarization for vectorization, which is suitable for resolution conversion. However, vibrations of line drawings or border lines may occur, and then vectorization efficiency decreases. In this paper, we propose subpixel binarization based on a directive one-dimensional filtering along a border region of homogeneous tone. A step function is adopted as the one-dimensional filter for suppression of the vibration.

CiNii
ベクター表現による文字や網点を含む画像符号化システム

河村, 山本, 石井, 渡辺

情報処理学会全国大会 3B-1 2007.03

CiNii
コミック画像におけるコマ分割処理の高速化に関する検討

石井, 河村, 渡辺

情報処理学会全国大会 1P-2 2007.03

CiNii
複雑背景における物体検出法

伊谷, 渡辺

電子情報通信学会総合大会 D-12-152 2007.03
照明変化のある環境下での移動物体検出の検討

谷, 伊谷, 渡辺, 富永

電子情報通信学会総合大会 D-11-102 102 - 102 2007.03

CiNii
複数画像によるダイナミックレンジ向上に関する検討

高木, 河村, 渡辺, 富永

電子情報通信学会総合大会 D-11-67 2007.03
縮小画像を用いたフレーム補間におけるギャップ処理手法に関する検討

東, 渡辺

電子情報通信学会総合大会 D-11-65 65 - 65 2007.03

CiNii
電子透かしを埋め込んだベクター画像の客観評価手法に関する検討

河村, 山本, 石井, 渡辺

電子情報通信学会総合大会 D-11-42 2007.03
方向性フィルタバンクを用いた多重記述符号化における係数補間に関する検討

石川, 渡辺

電子情報通信学会総合大会 D-11-32 32 - 32 2007.03

CiNii
時空間画像のJPEG 2000符号化によるPSNRの周期的変動の低減に関する検討

加藤, 渡辺

電子情報通信学会総合大会 D-11-28 2007.03
エッジ成分のベクトル化による空間スケーラブル符号化

山本, 河村, 渡辺

電子情報通信学会総合大会 D-11-8 2007.03
FGS映像における受信画質劣化防止に関する一検討

秋間, 渡辺, 富永

電子情報通信学会総合大会 D-11-7 2007.03
View Interpolation を用いた多視点符号化の効率改善に関する一検討

後藤, 内藤, 渡辺

電子情報通信学会総合大会 D-11-6 2007.03
複雑な背景における物体検出法の検討

伊谷, 渡辺

電子情報通信学会PRMU研究会 2006PRMU12-09 2006.12
FGSビデオ受信時におけるビットストリームの信頼性に関する検討

秋間, 渡辺, 富永

情報処理学会AVM研究会 2006AVM55-20 ( 67 ) 51 - 54 2006.12

CiNii
方向性フィルタバンクによる多重記述符号化に関する検討

石川, 渡辺

情報処理学会AVM研究会 2006AVM55-19 2006.12
複素ウェーブレットを用いた画像符号化における情報量削減手法に関する検討

高橋, 河村, 渡辺

画像符号化シンポジウム PCSJ2006 P-5.09 2006.11
画像の不確定性符号化について

石川, 渡辺

画像符号化シンポジウム PCSJ2006 P-3.02 2006.11

CiNii
コミックのコマ分割処理に関する一検討

石井, 河村, 渡辺

画像符号化シンポジウム PCSJ2006 P-2.07 2006.11
ベクター表現によるマンガ符号化システム

河村, 山本, 石井, 渡辺

画像符号化シンポジウム PCSJ2006 P-2.06 2006.11
奥行き情報を用いた多視点映像符号化の符号量削減

後藤, 内藤, 渡辺

画像符号化シンポジウム PCSJ2006 P-1.03 2006.11
動きオブジェクトベースのフレーム補間手法に関する検討

東, 渡辺

FIT2006 J-069 2006.09
分割線選択によるコミックのコマ分割処理に関する検討

石井, 河村, 渡辺

FIT2006 J-033 2006.09

CiNii
文字や網点を含む低解像度多値画像のベクター表現に関する検討

河村, 山本, 渡辺

FIT2006 J-032 261 - 262 2006.09

CiNii
複素ウェーブレット変換を用いた画像符号化に関する一検討

高橋, 河村, 渡辺

FIT2006 J-031 259 - 260 2006.09

CiNii
JPEG特許に関する最近の話題

渡辺

情報処理学会誌 Vol.47 ( No.9 ) 2006.09
SPORT: Extended Simulations and Results for Divisible Load Scheduling on Heterogeneous Systems

A. Ghatpande, H. Nakazato, H. Watanabe

IEICE Society Conference BS-15-3 2006.09
ベクター表現による文字や網点を含む画像符号化システム

河村, 山本, 石井, 渡辺

情報処理学会AVM研究会 2006AVM54-06 2006.09
イントラ符号化にベクター表現を用いた動画像圧縮に関する検討

河村, 山本, 渡辺

情報処理学会全国大会 2F-2 2006.03
ベクトル表現を用いた空間スケーラブル符号化

山本, 河村, 渡辺

電子情報通信学会総合大会 D-11-1 2006.03
映像シーンに適応したスケーラビリティ選択手法

高橋, 石川, 渡辺, 富永

電子情報通信学会総合大会 D-11-19 2006.03
フレーム補間におけるギャップ処理に関する検討

東, 渡辺, 永吉, 富永

電子情報通信学会総合大会 D-11-20 2006.03
カラー画像の特性を考慮した超解像手法に関する検討

早雨, 河村, 永吉, 渡辺, 富永

電子情報通信学会総合大会 D-11-92 2006.03
線画の混在する画像におけるテキスト領域の抽出に関する検討

河村, 山本, 渡辺

電子情報通信学会総合大会 D-11-99 2006.03
人物のふるまいに着目した動的特徴量からの人物照合に関する一検討

守屋, 河村, 渡辺, 富永

電子情報通信学会総合大会 D-12-96 2006.03
逆転音声による音声・音楽の半開示スクランブルに関する検討

本多, 河村, 渡辺, 富永

電子情報通信学会総合大会 D-14-15 2006.03
A study on spatial scalable coding using vector representation

Yuki Yamamoto, Kei Kawamura, Hiroshi Watanabe

2006 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO - ICME 2006, VOLS 1-5, PROCEEDINGS MA1-P2.4 145 - + 2006 [Refereed]

　View Summary

The major advantage of vector representation of an image is that the image quality is maintained for arbitrary scaling. In recent years, a demand for scalable image coding has been increasing because of the wide variety of available digital contents and display terminals. Conventional scalable coding schemes are based on raster representation, and thus, line drawings deteriorate in quality when expanded and shrunk. In this paper, we propose an edge reconstruction method using vector representation for the purpose of keeping a consistent spatial scalability on transmission and display. We take an anti-aliasing into account in edge areas for approximation of luminance values around the edge. The proposed method can improve PSNR by up to 2 dB as compared to the conventional methods when image is expanded and shrunk.
Bit rate reduction of vector representation of binary images

Yuki Yamamoto, Kei Kawamura, Hiroshi Watanabe

2006 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP 2006, PROCEEDINGS WP-P3.7 3105 - + 2006 [Refereed]

　View Summary

Vector representation of binary images has an advantage of keeping high image quality for arbitrary scaling as well as editing capability of an object. However, the vector representation suffers from low compression efficiency compared with JBIG. In this paper, we show the main cause reducing coding efficiency and propose two methods to improve it. The proposed methods can reduce the file size of a binary image up to about 30-40 percent.
ベクトル表現を用いた空間スケーラブル符号化に関する一検討

山本, 河村, 渡辺

情報処理学会AVM研究会 2005AVM51-14 75 - 80 2005.12
動画像圧縮におけるベクター表現の符号化に関する検討

河村, 山本, 渡辺

情報処理学会AVM研究会 2005AVM51-15 81 - 86 2005.12
品質変動低減のためのJPEG2000符号量制御方式

佐野, 内藤, 渡辺

情報処理学会AVM研究会 2005AVM51-17 93 - 98 2005.12
複雑な環境における物体検出法の一検討

伊谷, 渡辺

画像符号化シンポジウム PCSJ2005 2005.11
ベクトル表現を用いた時空間スケーラブル符号化におけるエッジ再構成に関する検討

山本, 河村, 渡辺

画像符号化シンポジウム PCSJ2005 2005.11
イントラ符号化にベクター表現を用いた動画像圧縮に関する検討

河村, 山村, 渡辺

画像符号化シンポジウム PCSJ2005 2005.11

CiNii
映像シーンに適応したスケーラビリティ選択手法に関する検討

高橋, 後藤, 石川, 渡辺, 富永

画像符号化シンポジウム PCSJ2005 P-5.13 2005.11
動き補償付時間方向フィルタリングにおける適応的Update 手法に関する一検討

後藤, 石川, 渡辺

画像符号化シンポジウム PCSJ2005 2005.11
時空間画像のJPEG 2000 符号化におけるSNR の周期的変動に関する考察

加藤, 石川, 渡辺

画像符号化シンポジウム PCSJ2005 P-2.18 2005.11
ディジタルシネマのフリッカー低減を目的とした符号化手法

石川, 渡辺

画像符号化シンポジウム PCSJ2005 P-2.17 2005.11

CiNii
H.264 Inter予測における動き補償ブロックサイズ高速決定手法

常松, 渡辺

情報処理学会AVM研究会 2005AVM50-3 ( 9 ) 1979 - 1981 2005.10

CiNii
ベクトル変換におけるパスの本数と点の個数に関する検討

河村, 山本, 渡辺

FIT2005 J-075 2005.09
時空間スケーラビリティを考慮したセルアニメーションにおけるエッジ再構成に関する一検討

山本, 河村, 渡辺

FIT2005 J-068 2005.09
JPEG2000によるHDTV低遅延符号化のための符号量制御方式

佐野, 内藤, 渡辺

FIT2005 J-062 2005.09
Motion JPEG 2000における前後フレームを用いたフリッカー低減手法の検討

伊谷, 渡辺

FIT2005 J-061 2005.09
時空間画像のJPEG 2000符号化におけるSNR揺らぎについて

加藤, 石川, 渡辺

FIT2005 J-060 2005.09
SPORT: A Near-Optimal Solution to Divisible Load Scheduling on Heterogeneous Systems

A.Ghatpande, H.Nakazato, H.Watanabe

2005 IEICE Society Conference BS-9-4 2005.09
H.264/AVC におけるIntra 予測モードの決定手法に関する検討

常松, 渡辺

電子情報通信学会論文誌D-II Vol. J88-D2 ( No.9 ) 1979 - 1981 2005.09

CiNii
空間スケーラビリティを用いた画像解像度変換法

石川, 渡辺

画像電子学会誌 Vol. 34 ( No. 5 ) 548 - 557 2005.09
Motion JPEG 2000における静動領域判定に基づくフリッカー低減手法の検討

伊谷, 渡辺

情報処理学会AVM研究会 2005AVM49-14 ( 164 ) 77 - 82 2005.07

CiNii
“フィルムグレインを考慮したH.264/AVC符号化方式に関する一検討

後藤, 常松, 渡辺

情報処理学会AVM研究会 2005AVM49-06 2005.07
H.264における動き補償ブロックサイズ高速決定手法に関する一検討

常松, 後藤, 渡辺

情報処理学会AVM研究会 2005AVM49-04 ( 66 ) 19 - 24 2005.07

　View Summary

The performance improvement in H.264/MPEG-4 AVC is provided at the expense of higher computational complexity. Most of the complexity is caused by Inter prediction. To improve coding efficiency, some functions are added in H.264/MPEG-4 AVC, such as variable block size motion compensation, multi reference frame and quarter-pel motion compensation. A fast macroblock partition decision method is proposed in this paper. The macroblock size is efficiently determined by using the pixel value difference between encoding and the referred macroblock.

CiNii
時空間画像の符号化に関する一考察

加藤, 渡辺, 安田

情報処理学会全国大会 3W5 2005.03
通過点削減によるベクトル表現の符号化効率改善

山本, 河村, 渡辺, 富永

情報処理学会全国大会 1L5 2005.03
ベクター変換における符号化効率の改善

河村, 山本, 渡辺

情報処理学会全国大会 3W4 2005.03
時空間画像の符号化による時間軸ノイズの低減に関する一考察

加藤, 渡辺, 安田

電子情報通信学会総合大会 D-11-67 2005.03
Motion JPEG 2000におけるフリッカー低減手法の検討

伊谷, 石川, 渡辺, 富永

電子情報通信学会総合大会 D-11-61 61 - 61 2005.03

CiNii
フィルムグレインを考慮したH.264/AVCモード決定手法に関する一検討

後藤, 常松, 渡辺, 富永

電子情報通信学会総合大会 D-11-56 56 - 56 2005.03

CiNii
モードの偏りを用いたH.264/AVC FRExt Intra モード決定手法

常松, 渡辺

電子情報通信学会総合大会 D-11-55 55 - 55 2005.03

CiNii
二値画像のベクトル表現における符号化効率改善

山本, 河村, 渡辺, 富永

電子情報通信学会総合大会 D-11-30 2005.03
ベクター表現における階層的変換手法の検討

河村, 山本, 渡辺

電子情報通信学会総合大会 D-11-6 6 - 6 2005.03

CiNii
Gradation approximation for vector based compression of comic images

K Kawamura, Y Yamamoto, H Watanabe

2005 International Conference on Image Processing (ICIP), Vols 1-5 WedAmPO6 2845 - 2848 2005 [Refereed]

　View Summary

In this paper, we propose a method to vectorize comic images including halftone dots. Our method can prevent the jaggy and moire phenomena when the images are enlarged and shrank. The method contains three modules: halftone dots separation, gradation approximation, and vectorization. At the first module, small isolated areas and altered areas by dilation and erosion operations are separated into halftone dots images. At the second module, areas of halftone are approximated by contours and gradation parameters. At the last module, both halftone areas and line drawings are vectorized and approximated by smooth contours. The size of compressed file produced by our method is equal or smaller than JBIG compression. Validity of the proposed method is confirmed by experimental results.
H.264/AVC FRExtにおけるIntra予測モードの決定手法に関する一検討

常松, 渡辺

情報処理学会AVM研究会 2004AVM47-25 ( 77 ) 55 - 60 2004.12

CiNii
Motion JPEG2000を利用したマルチ解像度変換システム

石川, 渡辺

情報処理学会AVM研究会 2004AVM47-25 2004.12
Motion JPEG2000低遅延符号化のための時間方向ビット配分

佐野, 内藤, 渡辺

情報処理学会AVM研究会 2004AVM47-24 ( 77 ) 43 - 48 2004.12

CiNii
ベクター表現の階層化に関する検討

河村, 山本, 渡辺

情報処理学会AVM研究会 2004AVM47-23 2004.12
Distributed Video Encoding on Single-level Heterogeneous Processor Trees

A.Ghatpande, H.Nakazato, H.Watanabe

画像符号化シンポジウム PCSJ2004 P-5.11 2004.11
二値画像のベクトル化における符号量削減に関する検討

山本, 河村, 渡辺, 富永

画像符号化シンポジウム PCSJ2004 P-2.04 2004.11

CiNii
H.264/AVCにおけるIntra予測モードの決定手法に関する一検討

常松, 渡辺

画像符号化シンポジウム PCSJ2004 P-5.16 2004.11

CiNii
Distributed Video Encoding on Single-level Heterogeneous Processor Trees

A.Ghatpande, H.Nakazato, H.Watanabe

画像符号化シンポジウム PCSJ2004 P-5.11 2004.11
ベクター表現における符号量制御に関する検討

河村, 山本, 渡辺

画像符号化シンポジウム PCSJ2004 P-5.04 2004.11

CiNii
JPEG2000 Precinct構造を用いた低遅延HDTV符号化方式

佐野, 内藤, 渡辺

情報処理学会AVM研究会 2004AVM46-02 2004.10

CiNii
ベクター変換における曲線最適化アルゴリズムの一検討

河村, 渡辺

FIT2004 J-025 2004.09

CiNii
JPEG2000 Precinct構造による低遅延HDTV符号化方式

佐野, 内藤, 渡辺

FIT2004 J-076 2004.09
網点を含む2値画像のベクトル表現による高能率符号化に関する検討

河村, 渡辺

画像電子学会年次大会 2004.06
スプライトを適応的に利用したMPEG-4映像符号化システム

秦泉寺, 岡田, 八島, 渡辺

画像電子学会誌 No.33, Vol.3 ( 3 ) 333 - 342 2004.06

　View Summary

MPEG-4, a new audio-visual coding standard has strong compression tools, such as sprite coding. We have focused on sprite coding and proposed “Sprite Mode”. In sprite mode, video sequence is divided into foreground object and background object. Foreground object is compressed using MPEG-4 object coding, while background object is compressed using sprite coding. Sprite mode does not always adopt any video sequence such that images having still camera motion can be compressed in enough quality in low bit-rate. So, in this paper, we developed new MPEG-4 video encoder using sprite coding. In the proposed MPEG-4 encoder, multi-mode coding is newly proposed. Video shot suitable for conventional coding method is coded by normal mode (conventional method), while video shot suitable for sprite coding is coded by sprite mode. These two modes are automatic selected in any video sequences. We improved the processing flow using stuck system in order to consequently process video such as AVI format. We improved global motion detection algorithm, foreground object and background sprite generation algorithm. Then we proposed the simple rate-control for sprite mode. Then, we have some experiment on MPEG-4 video coding. The result shows effectiveness' of proposed encoder in low bit-rate coding.<br>

DOI CiNii

Scopus
モバイル環境におけるユーザクラスタリングを用いた情報推薦システムの検討

金田, 渡辺

情報処理学会AVM研究会 2003AVM44-04 2004.03
マンガ符号化における網点の多階調近似処理の評価

河村, 渡辺, 富永

情報処理学会全国大会 3Z-1 2004.03
マンガ符号化における濃度勾配検出法の改良

河村, 渡辺, 富永

情報処理学会AVM研究会 2003AVM44-03 ( 25 ) 13 - 18 2004.03

　View Summary

Comic coding by gradation approximation and vector representation has been proposed. When gradation approximation is processed in conventional method, we used some small areas with constant size. However, in some cases, we could not detect correctly because of the moire. In this paper, we study a detection method using a period of halftone dots. Then, we propose an improved method for density-gradient detection. Our approach decreases occurrence of moire. Through some simulation experiments, we have confirmed that our proposed method is valid.

CiNii
ディジタルシネマにおける階層符号化に関する検討

石川, 田邉, 渡辺

電子情報通信学会総合大会 D-11-15 2004.03
Motion-JPEG2000の符号量制御のためのROI符号化に関する検討

佐野, 渡辺, 富永

電子情報通信学会総合大会 D-11-17 2004.03
Motion JPEG2000映像配信のためのマルチ解像度変換法

田邉, 石川, 渡辺

情報処理学会AVM研究会 2003AVM44-19 2004.03

CiNii
Motion JPEG2000におけるフリッカー雑音とタイリングの関係について

加藤, 石川, 渡辺

情報処理学会AVM研究会 2003AVM44-18 2004.03
2値画像のベクトル表現における網点の階調近似処理に関する検討

河村, 渡辺, 富永

電子情報通信学会総合大会 D-11-106 2004.03
Vector representation of binary images containing halftone dots

K Kawamura, H Watanabe, H Tominaga

2004 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXP (ICME), VOLS 1-3 TP2-1 335 - 338 2004 [Refereed]

　View Summary

Vector representation of graphics has an advantage of being displayed at any size. When resolution of a bitmap image is changed, lack of a line segment arises. In addition, moire occurs when resolution of an image with halftone dots is resized. In this paper, we propose a new technique to convert the binary image with halftone dots into the vector representation. Resolution conversion of binary image can easily be performed without moire by continuous tone approximation of halftone dots. First, we separate the area of halftone dots and line drawings in the image. Next, a continuous tone approximation is applied to the area of halftone dots. Then, the conventional vectorization is applied to both continuous tone areas and line drawings. Finally these components are mixed and reconstructed. Our approach provides an efficient way of displaying cartoon like images at any size with limited amount of data.
マンガ符号化における網点のグラデーション処理の検討

河村, 渡辺, 富永

情報処理学会AVM研究会 2003AVM43-8 2003.12
網点を含んだ2値画像のベクトル表現に関する検討

河村, 渡辺, 富永

画像符号化シンポジウム PCSJ2003 P-2.13 2003.11

CiNii
ROIを利用したMotionJPEG2000の符号量制御に関する検討

佐野, 渡辺, 富永

画像符号化シンポジウム PCSJ2003 P-2.14 39 - 40 2003.11

CiNii
Motion JPEG 2000における最適解像度変換法

田邉, 渡辺, 富永

画像符号化シンポジウム PCSJ2003 P-1.02 2003.11

CiNii
An Architecture for Distributed Video Encoding on the Internet

A.Ghatpande, H.Nakazato, H.Watanabe

画像符号化シンポジウム PCSJ2003 P-2.02 2003.11
Distributed computing for real-time video processing

H.Watanabe, A.Ghatpande, H.Nakazato

1st International Conference on Ubiquitous Computing (ICUC) 2003, 2003.10
マンガの超高圧縮符号化に関する検討

河村, 渡辺, 富永

情報処理学会AVM研究会 2003AVM42-2 ( 99 ) 7 - 16 2003.10

　View Summary

Vectorization is an effective technique for comic image coding, since comics are mainly consist of line drawings. However, an existence of halftone-dots causes an increase of coding bitrate if simple vectorization is used. In addition, moires occur when a resolution of the image with halftone-dots is changed. In this paper, we propose a new technique to achieve highly efficient comic image coding. First, we separate the area of halftone-dots and line drawings from an image. Then, a continuous tone approximation is applied to the area of halftone-dots. Next, the conventional vectorization is applied to line drawings. Finally, these two components are mixed together.

CiNii
Distributed computing for real-time video processing

H.Watanabe, A.Ghatpande, H.Nakazato

1st International Conference on Ubiquitous Computing (ICUC) 2003, 2003.10
大規模コンテンツ配信のためのアプリケーションレベル・マルチキャストツリー構築に関する検討

清水, 亀山, 渡辺

FIT2003 L-011 2003.09
モバイル環境におけるユーザの嗜好と位置を利用した情報推薦システムの検討

金田, 亀山, 渡辺

FIT2003 M-092 2003.09
Motion JPEG2000におけるフリッカー雑音と量子化誤差の関係について

加藤, 田邉, 渡辺, 富永

FIT2003 J-039 2003.09

CiNii
JPEG2000におけるテキスト分離符号化方式に関する検討

石川, 富永, 渡辺

FIT2003 J-036 2003.09
Transcoding by automatic ROI extraction from JPEG2000 bitstream

H.Watanabe, T.Ogita

PCS2003 WP1.4-7 2003.04
Transcoding by automatic ROI extraction from JPEG2000 bitstream

H.Watanabe, T.Ogita

PCS2003 WP1.4-7 2003.04
IDG基礎から学ぶ情報圧縮技術

亀山, 金子, 渡辺

(株)IDGジャパン ISBN4-87280-486-4 2003.04
差分画像を利用したアニメーション映像からのオブジェクト抽出

中神, 渡辺, 富永

情報処理学会AVM研究会 2003AVM40-06 2003.03
ホストの相対的位置情報を利用したコンテンツ配信に関する検討

清水, 亀山, 渡辺

電子情報通信学会総合大会 B-6-170 2003.03
Wavelet符号化を動画像に適用した際の視覚的歪みに関する検討

加藤, 田邉, 渡辺, 富永

電子情報通信学会総合大会 D-11-15 2003.03

CiNii
Motion JPEG2000における最適解像度変換法の検討

田邉, 渡辺, 富永

情報処理学会AVM研究会 2003AVM40-04 ( 24 ) 19 - 23 2003.03

　View Summary

Motion JPEG2000 is regarded as a promising coding system for an ultra high quality image like digital-cinema and HDTV. Wavelet transform used in Motion JPEG2000 enables to realize spatial scalability easily. However, the performance of the wavelet filter used in Motion JPEG2000 is worse than those of decimation filters used in image resampling and resizing. What is worse, the wavelet filter can only resize images to half resolution. We aim to provide of the best resampling method for Motion JPEG2000. In this paper, we investigate the performance of filters and its relationship with image quality.

CiNii
ディジタル放送教科書(上)- IDG情報通信シリーズMPEG-1/MPEG-2/ MPEG-4

(株)IDGジャパン ISBN4-87280-477-5 2003.02
Very low bit rate video compression using MPEG-4 sprite coding

K Jinzenji, H Watanabe, S Okada, N Kobayashi

ELECTRONICS AND COMMUNICATIONS IN JAPAN PART II-ELECTRONICS 86 ( 6 ) 83 - 93 2003 [Refereed]

　View Summary

The new MPEG-4 image coding standard that has recently been established offers higher image quality in low-bit-rate coding than conventional coding methods such as H.26X, MPEG-1, and MPEG-2, as well as coding for each object unit (a new function).
This paper proposes an algorithm for the automatic extraction of video objects consisting of a foreground object and a background sprite. It is based on "sprite coding," which can greatly reduce the number of code bits compared to conventional methods. The foreground and background video objects processed by the proposed method are subjected to MPEG-4 object coding and sprite coding, respectively, and are compared with MPEG-4 simple profile (normal) coding. In particular, the influence of the frame rate and the foreground object rate on the number of code bits was investigated. It was found that the proposed method can reproduce images with the same quality while using about one-fourth to one-half as many code bits as the normal coding method when the foreground rate is 10 to 15% of the whole image, regardless of the frame rate. In addition, the proposed method can attain more than double the frame rate of the normal coding method at a very low bit rate (e.g., 128 and 64 kbit/s) with a similar SNR. (C) 2003 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 86(6): 83-93, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjb.10158.

DOI

Scopus

1

Citation

(Scopus)
Very low bit rate video compression using MPEG-4 sprite coding

K Jinzenji, H Watanabe, S Okada, N Kobayashi

ELECTRONICS AND COMMUNICATIONS IN JAPAN PART II-ELECTRONICS 86 ( 6 ) 83 - 93 2003

　View Summary

The new MPEG-4 image coding standard that has recently been established offers higher image quality in low-bit-rate coding than conventional coding methods such as H.26X, MPEG-1, and MPEG-2, as well as coding for each object unit (a new function).
This paper proposes an algorithm for the automatic extraction of video objects consisting of a foreground object and a background sprite. It is based on "sprite coding," which can greatly reduce the number of code bits compared to conventional methods. The foreground and background video objects processed by the proposed method are subjected to MPEG-4 object coding and sprite coding, respectively, and are compared with MPEG-4 simple profile (normal) coding. In particular, the influence of the frame rate and the foreground object rate on the number of code bits was investigated. It was found that the proposed method can reproduce images with the same quality while using about one-fourth to one-half as many code bits as the normal coding method when the foreground rate is 10 to 15% of the whole image, regardless of the frame rate. In addition, the proposed method can attain more than double the frame rate of the normal coding method at a very low bit rate (e.g., 128 and 64 kbit/s) with a similar SNR. (C) 2003 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 86(6): 83-93, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/ecjb.10158.

DOI

Scopus

1

Citation

(Scopus)
A novel content distribution architecture utilizing network distance prediction

N Shimizu, W Kameyama, H Watanabe

7TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL, III, PROCEEDINGS Proc. Vol 13 - 17 2003 [Refereed]

　View Summary

Contents Delivery/Distribution Network (CDN) and Peer-to-Peer (P2P) network have been proposed for a large-scale contents distribution. We have been investigating what mechanisms are required for CDN, P2P, and both of them. It turns out that 1) optimal server selection and 2) content discovery are key role functions. In this paper, we propose a method for optimal server selection. The optimal server is defined to minimize end-to-end delay between a requesting host and the server. Here, end-to-end delay is regarded as network distance. Round trip time (RTT) can be used as a metric to predict end-to-end delay. However, it takes much cost to send a probe packet to a whole set of hosts in order to measure all RTT. Our proposed method send a probing packet to a set of servers called Landmarks, and predict which server is optimal based on these RTT. Simulation result shows the validity of the proposed method. We confirm that this approach can adapt to scalable architecture.
A novel content distribution architecture utilizing network distance prediction

N Shimizu, W Kameyama, H Watanabe

7TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL, III, PROCEEDINGS Proc. Vol 13 - 17 2003

　View Summary

Contents Delivery/Distribution Network (CDN) and Peer-to-Peer (P2P) network have been proposed for a large-scale contents distribution. We have been investigating what mechanisms are required for CDN, P2P, and both of them. It turns out that 1) optimal server selection and 2) content discovery are key role functions. In this paper, we propose a method for optimal server selection. The optimal server is defined to minimize end-to-end delay between a requesting host and the server. Here, end-to-end delay is regarded as network distance. Round trip time (RTT) can be used as a metric to predict end-to-end delay. However, it takes much cost to send a probe packet to a whole set of hosts in order to measure all RTT. Our proposed method send a probing packet to a set of servers called Landmarks, and predict which server is optimal based on these RTT. Simulation result shows the validity of the proposed method. We confirm that this approach can adapt to scalable architecture.
モバイルカメラを利用した地理情報提示方法に関する基礎検討

金田, 吉沢, 亀山, 渡辺

映像メディア処理シンポジウム IMPS2002 I-2.11 2002.11
アニメーション映像における動き情報検出の検討

中神, 渡辺, 富永

画像符号化シンポジウム PCSJ2002 P-4.03 2002.11
Wavelet変換符号化における量子化歪について

田邉, 加藤, 渡辺, 富永

画像符号化シンポジウム PCSJ2002 P-3.02 2002.11

CiNii
MPEG-4 very low bitrate video coding by adaptively utilizing sprite

K.Jinzenji, S.Okada, N.Kobayashi, H.Watanabe

IEEE International Conference on Multimedia Expo (ICME) WedAmPO3: Compression II ( 3 ) 333 - 342 2002.08

　View Summary

MPEG-4, a new audio-visual coding standard has strong compression tools, such as sprite coding. We have focused on sprite coding and proposed “Sprite Mode”. In sprite mode, video sequence is divided into foreground object and background object. Foreground object is compressed using MPEG-4 object coding, while background object is compressed using sprite coding. Sprite mode does not always adopt any video sequence such that images having still camera motion can be compressed in enough quality in low bit-rate. So, in this paper, we developed new MPEG-4 video encoder using sprite coding. In the proposed MPEG-4 encoder, multi-mode coding is newly proposed. Video shot suitable for conventional coding method is coded by normal mode (conventional method), while video shot suitable for sprite coding is coded by sprite mode. These two modes are automatic selected in any video sequences. We improved the processing flow using stuck system in order to consequently process video such as AVI format. We improved global motion detection algorithm, foreground object and background sprite generation algorithm. Then we proposed the simple rate-control for sprite mode. Then, we have some experiment on MPEG-4 video coding. The result shows effectiveness' of proposed encoder in low bit-rate coding.<br>

DOI CiNii

Scopus
MPEG-4 very low bitrate video coding by adaptively utilizing sprite

K.Jinzenji, S.Okada, N.Kobayashi, H.Watanabe

IEEE International Conference on Multimedia Expo (ICME) WedAmPO3: Compression II ( 3 ) 333 - 342 2002.08

　View Summary

MPEG-4, a new audio-visual coding standard has strong compression tools, such as sprite coding. We have focused on sprite coding and proposed “Sprite Mode”. In sprite mode, video sequence is divided into foreground object and background object. Foreground object is compressed using MPEG-4 object coding, while background object is compressed using sprite coding. Sprite mode does not always adopt any video sequence such that images having still camera motion can be compressed in enough quality in low bit-rate. So, in this paper, we developed new MPEG-4 video encoder using sprite coding. In the proposed MPEG-4 encoder, multi-mode coding is newly proposed. Video shot suitable for conventional coding method is coded by normal mode (conventional method), while video shot suitable for sprite coding is coded by sprite mode. These two modes are automatic selected in any video sequences. We improved the processing flow using stuck system in order to consequently process video such as AVI format. We improved global motion detection algorithm, foreground object and background sprite generation algorithm. Then we proposed the simple rate-control for sprite mode. Then, we have some experiment on MPEG-4 video coding. The result shows effectiveness' of proposed encoder in low bit-rate coding.<br>

DOI CiNii

Scopus
Dynamic ProgrammingとFeedback Algorithmによる閉曲線近似法

宮澤敏記, 中神央二, 渡辺裕, 富永英義

画像電子学会第30回年次大会 No.20 2002.06
物体形状の推定と表面質感に関する一考察

巽英之, 渡辺裕, 大谷淳

画像電子学会第30回年次大会 No.19 2002.06
二層構造によるアニメーション画像符号化の検討

中神央二, 宮澤敏記, 渡辺裕, 富永英義

画像電子学会第30回年次大会 No.15 2002.06
コンテントオリエンティッド符号化実現のためシステム提案

清水直人, 宮澤敏記, 亀山渉, 渡辺裕, 富永英義

画像電子学会第30回年次大会 No.14 2002.06
MPEG-2ビットストリームからのROI抽出とトランスコーダへの適用

田邉集, 泰泉寺久美, 渡辺裕, 富永英義

画像電子学会第30回年次大会 No.9 21 - 21 2002.06

　View Summary

Streaming of MPEG video on the network where various bandwidth are intermingled has been getting common. Thus, the research on transcoding, which changes the bit rate of video according to the bandwidth of a transmission line, has paid attention. In the conventional transcoding, degradation of image quality caused by bit rate reduction diffuse accross the whole image. This approach is not so efficient in picture coding system. We focus on realization of more efficient transcoding, which maintain the better image quality in ROI. The experiments we conducted in this paper are foreground extraction from MPEG-2 bitstream and image quality control within transcoder.

DOI CiNii
JPEG2000トランスコーダにおけるROI自動抽出と評価

荻田健夫, 渡辺裕

画像電子学会第30回年次大会 No.10 2002.06

CiNii
A Study on Extraction of Moving Objects Using Watershed Transform

S.Watanabe, K.Jinzenji, H.Watanabe

画像電子学会第30回年次大会 No.8 20 - 20 2002.06

　View Summary

In this paper, a method to extract moving objects on a pixel base for MPEG-4 “sprite coding” is proposed. Sprite coding is a form of object coding: it uses a unified panoramic background image derived from a sequence having a camera motion, and a foreground object as video object planes (VOP’s). The proposed algorithm utilizes background difference and watershed transformation to extract the foreground moving objects. We first apply background difference to generate a foreground candidate image. Then, watershed transformation is applied to this candidate image to extract the contours of the foreground objects. Moreover, we utilize the macroblock mask of the foreground objects that is generated by the conventional method to delete the unnecessary edges extracted by watershed transformation. Results given by our proposed method are more than twice as better than that of the conventional method.

DOI CiNii
平滑化操作によるアニメーション画像符号化の検討

中神央二, 宮澤敏記, 渡辺裕, 富永英義

電子情報通信学会総合大会 D-11-54 2002.03
平滑化操作によるアニメーション画像符号化の検討

中神央二, 宮澤敏記, 渡辺裕, 富永英義

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 2002AVM36-11 2002.03
デコーダダウンローダブルシステムにおけるフレキシブルデコーダ構成

清水直人, 宮澤敏記, 亀山渉, 渡辺裕, 富永英義

電子情報通信学会総合大会 D-11-119 2002.03
デコーダダウンローダブルシステムにおけるスケジューリング方式の検討

宮澤敏記, 清水直人, 亀山渉, 渡辺裕, 富永英義

電子情報通信学会総合大会 D-11-120 2002.03
スプライトを用いたMPEG-4超高圧縮符号化方式:VideoESPER

泰泉寺久美, 岡田重樹, 小林直樹, 渡辺裕, 米原紀子

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 2002AVM36-8 ( 25 ) 43 - 51 2002.03

　View Summary

MPEG-4, a new audio-visual coding standard has strong compression tools, such as sprite coding. We have focused on sprite coding and proposed "Sprite Mode". In sprite mode, video sequence is divided into foreground object and background object. Foreground object is compressed using MPEG-4 object coding., while background object is compressed using sprite coding. Sprite mode does not always adopt any video sequence such that images having still camera motion can be compressed in enough quality in low bit-rate. So, in this paper, we developed new MPEG-4 video encoder using sprite coding, VideoESPER , which means Video Efficient Sprite aided EncodeR. In Video ESPER, multi-mode coding is newly proposed. Video shot suitable for conventional coding method is coded by normal mode (conventional method), while video shot suitable for sprite coding is coded by sprite mode. These two modes are automatic selected in any video sequences. We improved the processing flow using stuck system in order to consequently process video such as AVI format. We improved global motion detection algorithm, foreground object and background sprite generation algorithm. Then we proposed the simple rate-control for sprite mode and geometric sprite transforming method for more efficient sprite coding. Then, we have some experiment on MPEG-4 video coding. The result shows effectiveness' of VideoESPER in low bit-rate coding. On the other hand, limitation of ViddeoESPER is also discussed.

CiNii
ROIの自動抽出に基づくJPEG2000トランスコーダの一検討

電子情報通信学会総合大会 D-11-118 2002.03

CiNii
MPEG-2 Bit Stream からの動領域抽出法の検討

田邉集, 泰泉寺久美, 渡辺裕, 富永英義

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 2002AVM36-7 2002.03
JPEG2000を用いた静止画中の物体の質感表現の検討

電子情報通信学会総合大会 D-11-172 2002.03
Dynamic Programmingによるセル画アニメーション内均等色輪郭線符号化の検討

MIYAZAWA Toshinori, NAKAGAMI Ouji, KAMEYAMA Wataru, WATANABE Hiroshi, TOMINAGA Hideyoshi

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 2002AVM36-12 ( 25 ) 71 - 76 2002.03

　View Summary

There are many homogeneous color regions in animation images. By approximating the regions by significant points and representative colors, coding efficiency can be improved. In this paper, we propose an approximation method for the regions based on dynamic programming. An approximation criterion used in the method is an error value defined by the distance between an original and an approximated line. If the error value is smaller than a threshold, a significant point that provides the minimum error value is extracted from the line. This process is iteratively executed until the error value exceeds the threshold. Since the the number of iteration tends to be large, we evaluate the processing time of significant elements, which is dominant for computational complexity. The remaining problems, in the proposed approximation, are 1) lack of point reduction ratio contorol and 2) excess extraction of significant points at some threshold. We also propose feedback algorithm in order to solve the problems.

CiNii
MPEG-4 very low bit-rate video compression by adaptively utilizing sprite to short sequences

Kumi Jinzenji, Shigeki Okada, Naoki Kobayashi, Hiroshi Watanahe

Proceedings - 2002 IEEE International Conference on Multimedia and Expo, ICME 2002 1 653 - 656 2002

　View Summary

In MPEG-4, a video sequence can be divided into foreground object and background objects that are independently encoded. Using sprites can dramatically compress the overall bit rate, but not all video sequences can be so encoded. This paper introduces MPEG-4 multimode coding
it offers automatic coding mode decision, video object generation and high compression efficiency. The source video sequence is segmented and each segment is automatically categorized as either "normal", which is encoded using MPEG-4 simple profile, or as "sprite". Coding experiments show that if the bit rate is low, multimode coding offers higher coding efficiency than regular MPEG-4 in terms of frame rate and image quality.

DOI
MPEG-4 very low bit-rate video compression by adaptively utilizing sprite to short sequences

Kumi Jinzenji, Shigeki Okada, Naoki Kobayashi, Hiroshi Watanahe

Proceedings - 2002 IEEE International Conference on Multimedia and Expo, ICME 2002 1 653 - 656 2002

　View Summary

In MPEG-4, a video sequence can be divided into foreground object and background objects that are independently encoded. Using sprites can dramatically compress the overall bit rate, but not all video sequences can be so encoded. This paper introduces MPEG-4 multimode coding
it offers automatic coding mode decision, video object generation and high compression efficiency. The source video sequence is segmented and each segment is automatically categorized as either "normal", which is encoded using MPEG-4 simple profile, or as "sprite". Coding experiments show that if the bit rate is low, multimode coding offers higher coding efficiency than regular MPEG-4 in terms of frame rate and image quality.

DOI
Pixel-based extraction of moving objects for sprite coding

S Watanabe, K Jinzenji, H Watanabe

6TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL III, PROCEEDINGS D-11-96 195 - 200 2002 [Refereed]

　View Summary

In this paper, a method to extract moving objects on a pixel base for MPEG-4 "sprite coding" is proposed. Sprite coding is a form of object coding: it uses a unified panoramic background image derived from a sequence having a camera motion, and a foreground object as video object planes (VOP's). The proposed algorithm utilizes background difference, which is the difference between the original image and the background image, and watershed transform to extract the foreground moving objects. We first apply background difference to generate a foreground candidate image. Then, watershed transform is applied to this candidate image to extract the contours of the foreground objects. Furthermore, we utilize the macroblock mask of the foreground objects, which is generated by the conventional method, to delete the unnecessary edges extracted by watershed transformation. Results given by our proposed method are more than twice as better than that of the conventional method.
A dynamic programming algorithm for downlink throughput optimisation in wideband CDMA systems

MK Gurcan, OO Oyefuga, S Kahn, BO Lijofi

THIRD INTERNATIONAL CONFERENCE ON 3G MOBILE COMMUNICATION TECHNOLOGIES No.20 ( 489 ) 484 - 488 2002 [Refereed]

　View Summary

This paper develops an approach for the maximisation of throughput in the downlink of third generation mobile systems, based on wideband CDMA. A mathematical model is outlined for a multi-service multiplexed environment, with constraints placed on the system, based on the class of service. The optimisation problem is stated in terms of the model parameters and adapted for use with a dynamic programming algorithm. Results are presented, showing a significant improvement in throughput when high loads are placed on the system, compared with a fixed bit-rate allocation.
Pixel-based extraction of moving objects for sprite coding

S Watanabe, K Jinzenji, H Watanabe

6TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL III, PROCEEDINGS Proc. Vol. III 195 - 200 2002

　View Summary

In this paper, a method to extract moving objects on a pixel base for MPEG-4 "sprite coding" is proposed. Sprite coding is a form of object coding: it uses a unified panoramic background image derived from a sequence having a camera motion, and a foreground object as video object planes (VOP's). The proposed algorithm utilizes background difference, which is the difference between the original image and the background image, and watershed transform to extract the foreground moving objects. We first apply background difference to generate a foreground candidate image. Then, watershed transform is applied to this candidate image to extract the contours of the foreground objects. Furthermore, we utilize the macroblock mask of the foreground objects, which is generated by the conventional method, to delete the unnecessary edges extracted by watershed transformation. Results given by our proposed method are more than twice as better than that of the conventional method.
A study on content-oriented coding scheme and decoder downloadable system

O Nakagami, N Shimizu, T Miyazawa, W Kameyama, H Watanabe, H Tominaga

6TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL III, PROCEEDINGS Proc. Vol. III 111 - 116 2002
Pixel-based extraction of moving objects for sprite coding

S Watanabe, K Jinzenji, H Watanabe

6TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL III, PROCEEDINGS Proc. Vol. III 195 - 200 2002 [Refereed]

　View Summary

In this paper, a method to extract moving objects on a pixel base for MPEG-4 "sprite coding" is proposed. Sprite coding is a form of object coding: it uses a unified panoramic background image derived from a sequence having a camera motion, and a foreground object as video object planes (VOP's). The proposed algorithm utilizes background difference, which is the difference between the original image and the background image, and watershed transform to extract the foreground moving objects. We first apply background difference to generate a foreground candidate image. Then, watershed transform is applied to this candidate image to extract the contours of the foreground objects. Furthermore, we utilize the macroblock mask of the foreground objects, which is generated by the conventional method, to delete the unnecessary edges extracted by watershed transformation. Results given by our proposed method are more than twice as better than that of the conventional method.
A study on content-oriented coding scheme and decoder downloadable system

O Nakagami, N Shimizu, T Miyazawa, W Kameyama, H Watanabe, H Tominaga

6TH WORLD MULTICONFERENCE ON SYSTEMICS, CYBERNETICS AND INFORMATICS, VOL III, PROCEEDINGS Proc. Vol. III 111 - 116 2002 [Refereed]
MPEG-4 very low bit-rate video compression by adaptively utilizing sprite to short sequences

K Jinzenji, S Okada, N Kobayashi, H Watanabe

IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I AND II, PROCEEDINGS WedAmPO3: Compression II 653 - 656 2002 [Refereed]

　View Summary

In MPEG-4, a video sequence can be divided into foreground object and background objects that are independently encoded. Using sprites can dramatically compress the overall bit rate but not all video sequences can be so encoded. This paper introduces MPEG-4 multi-mode coding; it offers automatic coding mode decision, video object generation and high compression efficiency. The source video sequence is segmented and each segment is automatically categorized as either "normal", which is encoded using MPEG-4 simple profile, or as "sprite". Coding experiments shows that if the bit rate is low, multi-mode coding offers more high coding efficiency than regular MPEG-4 in terms of frame rate and image quality.
A study on two-layer coding for animation images

O Nakagami, T Miyazawa, H Watanabe, H Tominaga

IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I AND II, PROCEEDINGS WedAmPO3: Compression II 621 - 624 2002 [Refereed]

　View Summary

A coding scheme specifically designed for animation images is proposed. Taking characteristics of animation images into account, lines and homogeneous color regions are extracted from animation images. Lines and contours of the homogeneous color regions are approximated by straightline and spline functions. We found that smoothing operations are effective to extract homogeneous regions from background images. Connected Filter is used for smoothing operation while keeping edges of the original image correctly. Precise approximation of the contours is also required. For this purpose, we propose to use Dynamic Programming with a feedback algorithm. As a result, an animation image can be represented by two layers; significant points as a base layer, and DCT blocks for an additional layer. For compensation scheme to the loss caused by smoothing operation, DCT is applied to the difference between the original images and lines and homogeneous color regions images. High-quality images can be obtained by adding this differential data to the approximated images.
MPEG-4 very low bit-rate video compression by adaptively utilizing sprite to short sequences

K Jinzenji, S Okada, N Kobayashi, H Watanabe

IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I AND II, PROCEEDINGS WedAmPO3: Compression II 653 - 656 2002 [Refereed]

　View Summary

In MPEG-4, a video sequence can be divided into foreground object and background objects that are independently encoded. Using sprites can dramatically compress the overall bit rate but not all video sequences can be so encoded. This paper introduces MPEG-4 multi-mode coding; it offers automatic coding mode decision, video object generation and high compression efficiency. The source video sequence is segmented and each segment is automatically categorized as either "normal", which is encoded using MPEG-4 simple profile, or as "sprite". Coding experiments shows that if the bit rate is low, multi-mode coding offers more high coding efficiency than regular MPEG-4 in terms of frame rate and image quality.
A study on two-layer coding for animation images

O Nakagami, T Miyazawa, H Watanabe, H Tominaga

IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA AND EXPO, VOL I AND II, PROCEEDINGS WedAmPO3: Compression II 621 - 624 2002

　View Summary

A coding scheme specifically designed for animation images is proposed. Taking characteristics of animation images into account, lines and homogeneous color regions are extracted from animation images. Lines and contours of the homogeneous color regions are approximated by straightline and spline functions. We found that smoothing operations are effective to extract homogeneous regions from background images. Connected Filter is used for smoothing operation while keeping edges of the original image correctly. Precise approximation of the contours is also required. For this purpose, we propose to use Dynamic Programming with a feedback algorithm. As a result, an animation image can be represented by two layers; significant points as a base layer, and DCT blocks for an additional layer. For compensation scheme to the loss caused by smoothing operation, DCT is applied to the difference between the original images and lines and homogeneous color regions images. High-quality images can be obtained by adding this differential data to the approximated images.
A novel decoder-downloadable system for content-oriented coding

N Shimizu, T Miyazawa, W Kameyama, H Watanabe, H Tominaga

GLOBECOM'02: IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE, VOLS 1-3, CONFERENCE RECORDS GEN-01-1 1638 - 1642 2002

　View Summary

In this paper, a new system architecture called decoder-downloadable system is described. The purpose of this system is to provide a uniform platform for the multimedia world based on image compressions using characteristics of the contents. This system enables dynamic decoder downloading by negotiation between servers and clients for seamless and minimum delay playback. This negotiation scheme is implemented by Java and CORBA (Common Object Request Broker). Thus, servers and clients in this system are independent from OSes and hardware specifications. This system is suitable for TV broadcasting as well as Internet streaming.
We improve a decoder architecture in order to make the system more flexible. A decoder has some functions such as bitstream parsing, image compression algorithms (DCT, Wavelet, Motion estimation and so on). However, conventional decoders are monolithic software. Thus, it is impossible to share parts of decoders. The proposed method enables a module-based decoder and sharing some modules among some decoders. Consequently, the flexible decoder can make downloading time shorter by avoidance to download redundant parts of decoders. It also achieves scalabillity enabling this system to be used in some multimedia applications, such as a multimedia content search system, as well as a multimedia player.
A novel decoder-downloadable system for content-oriented coding

N Shimizu, T Miyazawa, W Kameyama, H Watanabe, H Tominaga

GLOBECOM'02: IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE, VOLS 1-3, CONFERENCE RECORDS GEN-01-1 1638 - 1642 2002 [Refereed]

　View Summary

In this paper, a new system architecture called decoder-downloadable system is described. The purpose of this system is to provide a uniform platform for the multimedia world based on image compressions using characteristics of the contents. This system enables dynamic decoder downloading by negotiation between servers and clients for seamless and minimum delay playback. This negotiation scheme is implemented by Java and CORBA (Common Object Request Broker). Thus, servers and clients in this system are independent from OSes and hardware specifications. This system is suitable for TV broadcasting as well as Internet streaming.
We improve a decoder architecture in order to make the system more flexible. A decoder has some functions such as bitstream parsing, image compression algorithms (DCT, Wavelet, Motion estimation and so on). However, conventional decoders are monolithic software. Thus, it is impossible to share parts of decoders. The proposed method enables a module-based decoder and sharing some modules among some decoders. Consequently, the flexible decoder can make downloading time shorter by avoidance to download redundant parts of decoders. It also achieves scalabillity enabling this system to be used in some multimedia applications, such as a multimedia content search system, as well as a multimedia player.
Pull型サービスにおけるデコーダダウンローダブルシステムの構成及び評価

宮澤敏記, 清水直人, 亀山渉, 渡辺裕, 富永英義

画像符号化シンポジウム PCSJ2001 P-P2.06 2001.11
箱で運んだ会議ドキュメント

渡辺裕

画像電子学会誌 30;5;pp.629-630 2001.09
Sprite Coding in Object-based Video Coding Standard: MPEG-4

H.Watanabe, K.Jinzenji

World Multiconference on Systemics, Cybernetics and Informatics (SCI) 2001 Vol.XIII;pp.420-425 2001.07
Animation Image Coding

T.Miyazawa, W.Kameyama, H.Watanabe, T.Sakatani, H.Tominaga

World Multiconference on Systemics, Cybernetics and Informatics (SCI) 2001 Vol.XIII;pp.404-409 2001.07
Sprite Coding in Object-based Video Coding Standard: MPEG-4

H.Watanabe, K.Jinzenji

World Multiconference on Systemics, Cybernetics and Informatics (SCI) 2001 Vol.XIII;pp.420-425 2001.07
Animation Image Coding

T.Miyazawa, W.Kameyama, H.Watanabe, T.Sakatani, H.Tominaga

World Multiconference on Systemics, Cybernetics and Informatics (SCI) 2001 Vol.XIII;pp.404-409 2001.07
MPEG-4スプライト符号化を用いた超低レート動画像圧縮

秦泉寺久美, 渡辺裕, 岡田重樹, 小林直樹

電子情報通信学会論文誌D-II J84-D-II;5;pp.758-768 ( 5 ) 758 - 768 2001.05

CiNii
An Approach to MPEG-4 Multi Mode Coding Using Sprite Coding

S.Okada, K, Jinzenji, H.Watanabe, N.Kobayashi

International Picture Coding Symposium PCS2001 pp.421-424 2001.04
An Approach to MPEG-4 Multi Mode Coding Using Sprite Coding

S.Okada, K, Jinzenji, H.Watanabe, N.Kobayashi

International Picture Coding Symposium PCS2001 pp.421-424 421 - 424 2001.04

CiNii
アニメーション画像符号化の基礎検討

宮澤敏記, 亀山渉, 渡辺裕, 阪谷徹, 富永英義

電子情報通信学会 D-11-16 66 - 66 2001.03

CiNii
アニメーション画像符号化における輪郭線抽出と近似方式の検討

宮澤敏記, 亀山渉, 渡辺裕, 阪谷徹, 富永英義

情報処理学会AVM研究会 AVM32-12, pp. 65-70 2001.03
Information Technology:Multimedia Standards-what's next?

Hiroshi Watanabe

ISO Bulletin 32-1 page4 2001.01
スプライト符号化を利用したMPEG-4マルチモード符号化方式の研究

岡田重樹, 秦泉寺久美, 渡辺裕, 小林直樹

画像符号化シンポジウムPCSJ2000 P-P. 2. 05 2000.11

CiNii
MPEG-4, JPEG2000符号化の現在と将来像

画像符号化シンポジウムPCSJ2000 パネリストセッション 2000.11
Visual Communication in Mobile Era

MoMuC2000 パネリスト 2000.10
Visual Communication in Mobile Era

MoMuC2000 パネリスト 2000.10
SC29 and Multimedia

MoMuC2000 キーノートスピーカー 2000.10
MPEG-4スプライト符号化を用いた超低レート動画像圧縮

秦泉寺久美, 岡田重樹, 渡辺裕, 小林直樹

画像電子学会研究会 0-03-02;pp.7-13 7 - 13 2000.09

CiNii
Bitstream Scaling by Reference Frame Altering

清水淳, 嵯峨田淳, 渡辺裕, 小林直樹

電子情報通信学会論文誌D-(]G0002[) J83-D-II,3,967-976 ( 3 ) 967 - 976 2000.03

CiNii
背景スプライトを用いた動画像圧縮に関する一考察

Jinzenji Kumi, Okada Shigeki, Watanabe Hiroshi, Kobayashi Naoki

電子情報通信学会全国大会 D-11-52 52 - 52 2000.03

CiNii
参照フレーム変更によるビットストリームスケーリング

清水淳, 嵯峨田淳, 渡辺裕, 小林直樹

電子情報通信学会論文誌D-II J83-D-II,3,pp.967-976 2000.03

CiNii
スプライト符号化に適したショット判定方法

電子情報通信学会全国大会 D-11-53 2000.03
スプライト符号化に適したショットの判定方法

OKADA Shigeki, JINZENJI Kumi, WATANABE Hiroshi, KOBAYASHI Naoki

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 2000-AVM-28,28-11,pp.61-66 ( 24 ) 61 - 66 2000.03

　View Summary

Sprite coding is supported by the new video coding standard MPEG-4 Ver.1 Main Profile. It provides content-based functionality and low bit-rate video compression. We focused on this coding technique in order to develop high quality and very low bit-rate application that is effective in narrow-band transmission such as Internet. In this paper, we analyze that what kind of video sequence is suitable for sprite coding by checking a few parameters. This paper also proposes an automatic shot decision algorithm for sprite coding.

CiNii
Global Motion Estimation for Sprite Production and Application to Video Coding

JINZENJI K.

IEEE ISPACS'98 J83-D-II,2,535-544 328 - 332 2000.02

CiNii
スプライト生成のためのグローバルモーション算出法と符号化への適用

秦泉寺久美, 渡辺裕, 小林直樹

電子情報通信学会論文誌D-II J83-D-II;2,pp.535-544 ( 2 ) 535 - 544 2000.02

CiNii
スプライト生成のためのグローバルモーション算出法と符号化への適用

JINZENJI Kumi, WATANABE Hiroshi, KOBAYASHI Naoki

電子情報通信学会論文誌D-II J83-D-II;2,pp.535-544 ( 2 ) 535 - 544 2000.02

CiNii
Automatic two-layer video object plane generation scheme and its application to MPEG-4 video coding

K Jinzenji, S Okada, H Watanabe, N Kobayashi

ISCAS 2000: IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS - PROCEEDINGS, VOL III pp.606-609 606 - 609 2000 [Refereed]

　View Summary

The new video coding standard MPEG-4 provides content-based functionality and low bit-rate video compression. We focused on the "sprite coding" supported MPEG-4 Version 1 Main profile in order to achieve "VHS quality video on 2B (128bps)" for narrow-band transmission such as the Internet. Automatic VOP (Video Object plane) generation technologies are being studied as one of the most important issues of MPEG-4 object coding. This paper proposes a two-layer VOP generation scheme with some core algorithms such as Gh IE (Global Motion Estimation). foreground moving object extraction, and background sprite generation. This paper also describes a shape information reduction method for foreground objects. Using this method, shape information is compressed by 90%. Experiments are conducted on VOP generation and video coding with MPEG-4. Coding efficiency is 3-4 times higher than that of typical existing video coding schemes at the same subjective image quality.
Automatic two-layer video object plane generation scheme and its application to MPEG-4 video coding

K Jinzenji, S Okada, H Watanabe, N Kobayashi

ISCAS 2000: IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS - PROCEEDINGS, VOL III pp.606-609 606 - 609 2000

　View Summary

The new video coding standard MPEG-4 provides content-based functionality and low bit-rate video compression. We focused on the "sprite coding" supported MPEG-4 Version 1 Main profile in order to achieve "VHS quality video on 2B (128bps)" for narrow-band transmission such as the Internet. Automatic VOP (Video Object plane) generation technologies are being studied as one of the most important issues of MPEG-4 object coding. This paper proposes a two-layer VOP generation scheme with some core algorithms such as Gh IE (Global Motion Estimation). foreground moving object extraction, and background sprite generation. This paper also describes a shape information reduction method for foreground objects. Using this method, shape information is compressed by 90%. Experiments are conducted on VOP generation and video coding with MPEG-4. Coding efficiency is 3-4 times higher than that of typical existing video coding schemes at the same subjective image quality.
二層ビデオオブジェクト自動生成とMPEG-4符号化への適用

JINZENJI Kumi, WATANABE Hiroshi, KOBAYASHI Naoki

電子情報通信学会画像工学研究会 IE99-115,pp.49-54 ( 80 ) 49 - 54 1999.12

　View Summary

The new video coding standard MPEG-4 provides content-based functionality and low bit-rate video compression. We focused on the "sprite coding" supported MPEG-4 Version 1 Main profile in order to achieve "VHS quality video on 2B(128kbps)" for narrow-band transmission such as the Internet. Automatic VOP(Video Object Plane)generation technologies are being studied as one of the most important issues of MPEG-4 object coding. This paper proposes a two-layer VOP generation scheme with some core algorithms such as GME(Global Motion Estimation), foreground moving object extraction, and background sprite generation. This paper also describes a shape information reduction method for foreground objects. Using this method, shape information is compressed by 90%. Experiments are conducted on VOP generation and video coding with MPEG-4. Coding efficiency is several times higher than that of typical existing video coding schemes at the same objective image quality.

DOI CiNii
スプライン/DCTハイブリッド符号化方式を用いたアニメーション映像における領域抽出の検討

映像情報メディア学会マルチメディア情報処理研究会 MIP99-76,pp.23-30 1999.11
Next-generation multimedia coding standard: MPEG-4

H Watanabe, S Takamura, T Moriya

NTT REVIEW 11 ( 6 ) 73 - 74 1999.11 [Refereed]

　View Summary

PEG-LC is an international standard for multimedia coding that has been defined by ISO/IEC JTC 1/SC 29/WG 11. Conventional audio-visual coding is for compression of natural images, natural audio and speech, whereas, MPEG-LC is for compression of synthetic images such as computer graphics and synthetic audio as well as natural audio-visual sources. These sources are separately coded as different objects at the encoder so that they can be decoded according to user preference. This article describes MPEG-4 version one, for which the Final Draft International Standard (FDIS) voting phase has finished (1999).
Video Coding Using Global Motion and Brightness-variation Compensation with Complexity Reduction

KAMIKURA Jazuto, WATANBE Hiroshi, KOBAYASHI Naoki, ICHINOSE Susumu, YASUDA Hiroshi

The Transactions of the Institute of Electronics,Information and Communication Engineers B J82-B/9,1676-1688 ( 9 ) 1676 - 1688 1999.09

CiNii
演算量低減を考慮したグローバル動き・輝度変化補償動画像符号化

KAMIKURA Jazuto, WATANBE Hiroshi, KOBAYASHI Naoki, ICHINOSE Susumu, YASUDA Hiroshi

電子情報通信学会論文誌B J82-B;9,pp.1676-1688 ( 9 ) 1676 - 1688 1999.09

CiNii
演算量低減を考慮したグローバル動き・輝度変化補償動画像符号化(共著)

KAMIKURA Jazuto, WATANBE Hiroshi, KOBAYASHI Naoki, ICHINOSE Susumu, YASUDA Hiroshi

電子情報通信学会論文誌B J82-B/9,1676-1688 ( 9 ) 1676 - 1688 1999.09

CiNii
アニメーション映像における領域抽出の一検討

画像符号化シンポジウム(PCSJ '99) P.3.15,pp.65-66 65 - 66 1999.09

CiNii
VOP自動生成とMPEG-4符号化への適用

画像符号化シンポジウム(PCSJ '99) P.4.03,pp.73-74 1999.09
MPEG-2ビデオ符号化チップSuperENCにおける1-passVBR制御方式

画像符号化シンポジウム(PCSJ '99) P.1.02,pp.5-6 1999.09
次世代マルチメディア符号化国際標準MPEG-4

NTT技術ジャーナル 11;8,pp.56-57 1999.08
Automatic VOP Production Scheme for Very Low Bit Rate Video Coding

Picture Coding Symposium '99 pp.299-302 1999.04
Automatic VOP Production Scheme for Very Low Bit Rate Video Coding

JINZENJI K.

Picture Coding Symposium '99 pp.299-302 299 - 302 1999.04

CiNii
Global brightness-variation compensation for video coding

K Kamikura, H Watanabe, H Jozawa, H Kotera, SN Ichinose

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 8 ( 8 ) 988 - 1000 1998.12 [Refereed]

　View Summary

In this paper, a global brightness-variation compensation (GBC) scheme is proposed to improve the coding efficiency for video scenes that contain global brightness variations caused by fade in/out, camera-iris adjustment, flicker, illumination change, and so on. In this scheme, a set of tno brightness-variation parameters, which represent multiplier and offset components of the brightness variation in the whole frame, is estimated. The brightness variation is then compensated using the parameters. We also propose a method to apply the GBC scheme to component signals for video coding. Furthermore, a block-by-block ON/OFF control method for GBC is introduced to improve coding performance even for scenes including local variations in brightness caused by camera Bashes, spotlights, and the like, Simulation results show that the proposed CBC scheme with the ON/OFF control method improves the peak signal-to-noise ratio (PSNR) by 2-4 dB when the coded scenes contain brightness variations.
Moving Object Extraction Using Sum of Difference throughout Video Clip

Jinzenji Kumi, Watanabe Hiroshi, Kobayashi Naoki

PROCEEDINGS OF THE ITE ANNUAL CONVENTION 1998 415 1998

DOI CiNii
A Dynamic Computation Resource Scalable Algorithm for Software Video CODEC

J80-D-II,2,444-458 1997.02
動的演算量スケーラブルアルゴリズムによるソフトウェア画像符号化(共著)

電子情報通信学会論文誌D-(]G0002[) J80-D-II,2,444-458 1997.02
Obtaining higher resolution in Asia: Advances in digital signal processing for HDTV

Hiroshi Watanabe, Hiroshi Yasuda

IEEE Signal Processing Magazine 14 ( 4 ) 52 - 64 1997

DOI

Scopus
64 kbit/s Video coding algorithm using adaptive gain/shape vector quantization

Hiroshi Watanabe, Yutaka Suzuki

Signal Processing: Image Communication 1 ( 2 ) 87 - 102 1989

　View Summary

A hybrid coding algorithm, consisting of motion compensated interframe prediction and adaptive gain/shape vector quantization, is proposed for low bit rate video coding. A video coding system using the proposed coding algorithm and its coding characteristics are also described. Motion compensated interframe prediction is an efficient technique for reducing temporal redundancy contained in moving pictures. In the proposed scheme, constant amplitude blocks ("gray level blocks") are also included in search vectors in order to increase coding efficiency for quick object movement or scene change. Adaptive gain/ shape vector quantization with a tree search codebook is employed to encode motion compensated interframe difference signals. It can respond to changes in source statistics because its codebook is independent of the gain components of the input vectors. The proposed coding algorithm has been evaluated by computer simulations and shown to be effective for low bit rate video transmission. A 64 kbit/s video coding system based on this coding algorithm has been implemented. The developed coding system can transmit multiplexed video, audio and digital data at the basic ISDN rate, and can be applied to videophone as well as videoconferencing. © 1989.

DOI

Scopus

4

Citation

(Scopus)
MPEG-2 Coding Control Technology

NTT R&D 46/6,589-594
MPEG2符号化制御技術(共著)

NTT R&D 46/6,589-594

▼display all

Books and Other Publications

そこが知りたい最新技術オーディオ・ビデオ圧縮入門

亀山, 金子, 渡辺

(株)インプレスR 2007.04
改訂版ディジタル放送教科書(上)(下)

亀山, 花村, 渡辺ほか

(株)インプレス 2004.10 ISBN: 4844320270
Fundamentals of Digital Image Compression

1996
ディジタル画像圧縮の基礎(共著)

日経BP社 1996

Research Projects

Improving the accuracy of human motion analysis by machine learning

Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research

Project Year :

2020.04

-

2023.03
Research on Automatic Digest Creation for Large Scale Comic Database

Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research

Project Year :

2017.04

-

2020.03

WATANABE HIROSHI

　View Summary

In this study, we developed a method for detecting manga objects and automatically generating manga summaries using deep learning techniques to realize user-driven advanced manga search. We found that our method using Faster R-CNN provides the highest detection accuracy for detecting frames, character faces, speech balloons and texts of metadata. For image summarization, it was found that DBSCAN was effective in clustering the images of the character faces after detection.
Finally, automatic summaries were generated and their effectiveness was confirmed by subjective evaluation experiments. The proposed method was inadequate in terms of satisfaction and information content of the summaries, but the results were satisfactory in terms of comprehension.
Research on Automatic Digest Generation for Comic Retrieval

Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research

Project Year :

2013.04

-

2016.03

Watanabe Hiroshi, ISHI Daisuke

　View Summary

It is important to extract metadata contained in comics for realizing automatic digest generation system. The most important metadata is comic characters. In this research, we focused on the improvement of comic character extraction. Through experiments, we showed that Deformable Part Model (DPM), which is based on HOG features, is effective for extraction. We also studied on main character identification without prior knowledge by clustering. By this approach, positive and negative samples can be fed into DPM. Further, "Role Community Model", which is the method to identify the weight of each page by counting locations of character's appearance and their relations, can be used to realize automatic comic digest
マンガ検索のための自動要約生成に関する研究

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))

Project Year :

2013

-

2015

　View Summary

研究目的：マンガエピソードの自動要約生成には、登場キャラクターリストの自動生成が必須である。キャラクターの抽出には、その顔画像抽出が最も有効である。したがって、自然画像からの顔画像抽出を目的として開発されてきたハール変換型(Haar-like)の特徴量を、線画像主体で描かれるマンガ画像に対しても動作するように修正・拡張を行い、抽出精度の向上を目指している。
研究方法：キャラクター抽出を目的とした特徴量として、対象をハール変換型(Haar-like)のみに限定せず、勾配ヒストグラム(HOG)を基盤としたDeformable Part Modelなどの最新の手法も線画像に適用してその効果を確認した。また判別にはサポートベクターマシーン(SVM)を用いることとした。
キャラクター抽出には、マンガのキャラクター顔画像を数多く集積したデータセットが必要となる。そこで研究に協力的なマンガ作家の好意により、いくつかのマンガを提供してもらい顔画像（正例）切出し及び不正解画像（負例）切出しを行い、データベース化に向けた作業に入った。
研究成果：1. 勾配ヒストグラム(HOG)特徴量を用いてキャラクター抽出を試みる場合、正例と負例画像の識別の際にキャラクターが含まれる背景画像も抽出性能に影響を与えることが分かった。背景がセリフ入りの吹き出し画像、オノマトペが含まれる画像、それ以外に3分類した場合、オノマトペが背景に存在する場合には、キャラクター抽出への影響は小さいことが分かった。（電子情報通信学会総合大会において発表）
2. 勾配ヒストグラム(HOG)を基盤としたDeformable Part Modelは、顔画像中のいくつかの部分の相対的位置関係を利用するため、ハール変換型(Haar-like)特徴量よりもマンガキャラクターの顔画像抽出精度が高いことが分かった。（IEVC2014に投稿中）
Object-based Video Coding

Project Year :

2000

-

2001
Animation Image Coding

Project Year :

2000

-

▼display all

Misc

A Study for Video Coding Efficiency Using Static Sprite

Jinzenji Kumi, Watanabe Hiroshi, Kobayashi Naoki

Proceedings of the Society Conference of IEICE 1998 143 1998.09

CiNii
Grobal Motion Estimation for Static Sprite Production And Its Application to Video Coding

JINZENJI Kumi, WATANABE Hiroshi, ISHIBASHI Satoshi, KOBAYASHI Naoki

Technical report of IEICE. PRMU 98 ( 206 ) 47 - 53 1998.07

　View Summary

In static sprite coding, accuracy of produced static sprite has a strong effect on coding efficiency, SNR and subjectivity. Grobal Motion Estimation (GME) which is not sensitive to outlier data is desired to produce a stable static sprite from video sequence. In this paper, new GME which is not sensitive to outlier data is proposed. Then, using the proposed GME, coding efficiency and decoded image quality are evaluated in (1)Local motion compensation video coding, (2)GMC coding, (3)dynamic sprite coding and (4)static sprite coding of MPEG-4VM Ver.11. Consequently, static sprite coding achieved the highest coding efficiency among them.

CiNii
Grobal Motion Estimation for Static Sprite Production And Its Application to Video Coding

JINZENJI Kumi, WATANABE Hiroshi, ISHIBASHI Satoshi, KOBAYASHI Naoki

IEICE technical report. Image engineering 98 ( 204 ) 47 - 53 1998.07

　View Summary

In static sprite coding, accuracy of produced static sprite has a strong effect on coding efficiency, SNR and subjectivity. Grobal Motion Estimation (GME) which is not sensitive to outlier data is desired to produce a stable static sprite from video sequence. In this paper, new GME which is not sensitive to outlier data is proposed. Then, using the proposed GME, coding efficiency and decoded image quality are evaluated in (1)Local motion compensation video coding, (2)GMC coding, (3)dynamic sprite coding and (4)static sprite coding of MPEG-4VM Ver.11. Consenquently, static sprite coding achieved the highest coding efficiency among them.

CiNii
Grobal Motion Estimation for Static Sprite Production And Its Application to Video Coding

JINZENJI Kumi, WATANABE Hiroshi, ISHIBASHI Satoshi, KOBAYASHI Naoki

Technical report of IEICE. Multimedia and virtual environment 98 ( 208 ) 47 - 53 1998.07

　View Summary

In static sprite coding, accuracy of produced static sprite has a strong effect on coding efficiency, SNR and subjectivity. Grobal Motion Estimation (GME) which is not sensitive to outlier data is desired to produce a stable static sprite from video sequence. In this paper, new GME which is not sensitive to outlier data is proposed. Then, using the proposed GME, coding efficiency and decoded image quality are evaluated in (1) Local motion compensation video coding, (2) GMC coding, (3) dynamic sprite coding and (4) static sprite coding of MPEG-4VM Ver.11. Consequently, static sprite coding achieved the highest coding efficiency among them.

CiNii

Industrial Property Rights

画像符号化方法、画像符号化装置および画像符号化プログラム

5561611

渡辺裕, 石川孝明

Patent
画像符号化方法、画像符号化装置および画像符号化プログラム

5626728

石川孝明, 渡辺裕

Patent
画像符号化装置、画像符号化方法および画像符号化プログラム

5035926

渡辺裕, 石川孝明

Patent
画像符号化装置、画像符号化方法および画像符号化プログラム

5076150

渡辺裕, 石川孝明

Patent

Syllabus

Project Research B

School of Fundamental Science and Engineering

2024 fall semester
Multimedia Systems B

School of Fundamental Science and Engineering

2024 fall semester
Project Research A

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis B（Spring Semester）

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis A（Fall Semester）

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis B

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis A

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis B

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis B（Spring Semester）

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis A（Fall Semester）

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis B [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis A [S Grade]

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis A

School of Fundamental Science and Engineering

2024 spring semester
Computer Science and Engineering Laboratory B [S Grade]

School of Fundamental Science and Engineering

2024 spring semester
Computer Science and Engineering Laboratory A [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Computer Science and Engineering Laboratory B

School of Fundamental Science and Engineering

2024 spring semester
Computer Science and Engineering Laboratory A (2)

School of Fundamental Science and Engineering

2024 fall semester
Communications and Computer Engineering Laboratory A

School of Fundamental Science and Engineering

2024 fall semester
Project Research B

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis B

School of Fundamental Science and Engineering

2024 fall semester
Project Research A

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis B (Spring Semester)

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis B [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis A (Fall Semester)

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis A

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis A [S Grade]

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis A

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis A (Fall Semester)

School of Fundamental Science and Engineering

2024 fall semester
Bachelor Thesis B (Spring Semester)

School of Fundamental Science and Engineering

2024 spring semester
Bachelor Thesis B

School of Fundamental Science and Engineering

2024 fall semester
Multimedia Systems B [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Multimedia Systems B

School of Fundamental Science and Engineering

2024 fall semester
Computer Science and Communications Engineering Laboratory [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Computer Science and Communications Engineering Laboratory

School of Fundamental Science and Engineering

2024 fall semester
Communications and Computer Engineering Laboratory B [S Grade]

School of Fundamental Science and Engineering

2024 spring semester
Communications and Computer Engineering Laboratory B

School of Fundamental Science and Engineering

2024 spring semester
Communications and Computer Engineering Laboratory A [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Computer Science and Communications Engineering Laboratory B

School of Fundamental Science and Engineering

2024 spring semester
Graduation Thesis A (Spring) [S Grade]

School of Fundamental Science and Engineering

2024 spring semester
Project Research Spring

School of Fundamental Science and Engineering

2024 spring semester
Project Research Fall

School of Fundamental Science and Engineering

2024 fall semester
Multimedia Systems B

School of Fundamental Science and Engineering

2024 fall semester
Introduction to Computers and Networks

School of Fundamental Science and Engineering

2024 spring semester
Graduation Thesis B (Fall) [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Image Processing

School of Fundamental Science and Engineering

2024 spring semester
Graduation Thesis A (Fall) [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Graduation Thesis A (Spring)

School of Fundamental Science and Engineering

2024 spring semester
Computer Science and Communications Engineering Laboratory A [S Grade]

School of Fundamental Science and Engineering

2024 fall semester
Graduation Thesis A (Fall)

School of Fundamental Science and Engineering

2024 fall semester
Graduation Thesis B (Spring) [S Grade]

School of Fundamental Science and Engineering

2024 spring semester
Graduation Thesis B (Fall)

School of Fundamental Science and Engineering

2024 fall semester
Graduation Thesis B (Spring)

School of Fundamental Science and Engineering

2024 spring semester
Computer Science and Communications Engineering Laboratory A

School of Fundamental Science and Engineering

2024 fall semester
Research on Audiovisual Information Processing

Graduate School of Fundamental Science and Engineering

2024 full year
Special Laboratory B in Computer Science and Communications Engineering

Graduate School of Fundamental Science and Engineering

2024 fall semester
Seminar on Audiovisual Information Processing D

Graduate School of Fundamental Science and Engineering

2024 fall semester
Seminar on Audiovisual Information Processing C

Graduate School of Fundamental Science and Engineering

2024 spring semester
Seminar on Audiovisual Information Processing B

Graduate School of Fundamental Science and Engineering

2024 fall semester
Seminar on Audiovisual Information Processing A

Graduate School of Fundamental Science and Engineering

2024 spring semester
Research on Audiovisual Information Processing

Graduate School of Fundamental Science and Engineering

2024 full year
Special Laboratory A in Computer Science and Communications Engineering

Graduate School of Fundamental Science and Engineering

2024 spring semester
Multimedia and Standards

Graduate School of Fundamental Science and Engineering

2024 fall semester
Seminar on Audiovisual Information Processing D

Graduate School of Fundamental Science and Engineering

2024 fall semester
Seminar on Audiovisual Information Processing C

Graduate School of Fundamental Science and Engineering

2024 spring semester
Seminar on Audiovisual Information Processing B

Graduate School of Fundamental Science and Engineering

2024 fall semester
Seminar on Audiovisual Information Processing A

Graduate School of Fundamental Science and Engineering

2024 spring semester
Special Laboratory B in Computer Science and Communications Engineering

Graduate School of Fundamental Science and Engineering

2024 fall semester
Special Laboratory A in Computer Science and Communications Engineering

Graduate School of Fundamental Science and Engineering

2024 spring semester
Multimedia and Standards

Graduate School of Fundamental Science and Engineering

2024 fall semester
Special Seminar B in Computer Science and Communications Engineering

Graduate School of Fundamental Science and Engineering

2024 fall semester
Research on Audiovisual Information Processing

Graduate School of Fundamental Science and Engineering

2024 full year
Special Seminar A in Computer Science and Communications Engineering

Graduate School of Fundamental Science and Engineering

2024 spring semester

▼display all

Sub-affiliation

Faculty of Science and Engineering Graduate School of Fundamental Science and Engineering
Affiliated organization Global Education Center

Research Institute

2022

-

2024

Waseda Research Institute for Science and Engineering Concurrent Researcher
2022

-

2024

Global Information and Telecommunication Institute Concurrent Researcher

Internal Special Research Projects

敵対的生成ネットワークを用いた映像・音楽生成の高精度化

2020 Ahmad Yunnis Mousssa

　View Summary

Ahmad Yunis Moussa and Hiroshi Watanabe: “Audio Translation with Conditional Generative Adversarial Networks,” IEEE The 2nd International Conference on Artificial Intelligence in Information and Communication (ICAIIC 2020), 9B-1, Feb. 2020を研究基盤として、深層学習のひとつである敵対的生成ネットワークを用いて音響信号を合成する際の音質改善を図った。
機械学習による人物運動解析の高精度化

2019 Khin Sabai Htwe, 稲田健太郎, 大澤遼平, 山川敦也

　View Summary

本研究では，映像のみから人物の姿勢情報を機械学習により推定し，推定された人物の部位座標ベクトルを目的に応じて処理することにより，動作認識や個人認識の識別率を向上させることに成功した．体操競技において倒立状態の姿勢推定を可能にした．ダンスのような集団行動における動作タイミングの一致度を推定する手法について提案した．さらに，手話解析への適用を行い，映像上で手の重なりがない条件下での認識率を改善した．また，野球における投球動作解析のように同一シーンの整合が必要となる場合に，推定された部位座標を基に特徴ベクトルを生成することで，時系列整合アルゴリズムを用いて同一シーンの自動抽出を行えることを示した．
マンガ要約生成の自動化に関する研究

2016 柳澤秀彰

　View Summary

本研究では，Faster R-CNN(Regions with CNN features)を使用したマンガコンテンツの検出およびコマ内容の認識を提案した．コマやフキダシ，キャラクターといったオブジェクトは作品毎の形状変化が大きいため，既定の画像特徴量を用いた検出は困難である．従って，畳み込みニューラルネットワーク(CNN: Convolutional Neural Network)によって生成される特徴量のマンガへの適用について検討した．また，マンガコンテンツに固有な物体の配置に対応するため，複数の1クラス検出器の組み合わせによる検出手法を提案した．実験結果では，未知の5作品のマンガ画像に対する検出率とコマ構造認識率の観点から提案手法を評価した．その結果，52.8 - 84.9%の精度でコマ構造の認識が可能であることが分かった．
顔画像認識に基づいた動物の品種識別

2016 渡部宏樹

　View Summary

本研究では，動物の顔画像に基づく品種識別性能向上のために，DCGANを用いたCNNのデータオーギュメンテーション手法を提案した．実験では，猫および犬の顔画像を対象とし，提案手法の性能を品種識別の正解率により評価した．その結果，DCGANを用いた類似画像追加と従来の画像変換によるデータオーギュメンテーション手法の組み合わせにより，猫の場合で0.8%，犬の場合で1.3%の正解率の向上を確認した．さらに，生成画像の品質が提案手法に及ぼす影響についての調査を行った．実験の結果，品質の良い画像を加える手法の方がより高い性能を発揮することが分かった．
超高精細ビデオ符号化のための動きベクトル検出の研究

2013

　View Summary

本研究では，動き推定誤差データの曲面近似に基づく画素補間が不要な小数画素精度動き推定手法を提案した．実験結果では，提案手法の性能をPSNRとビットレートの観点から評価した．その結果，MPEG-4 AVC (Advanced Video Coding) 及びMPEG-4 HEVC (High Efficiency Video Coding) のいずれの符号化方式と組み合わせても，R-D性能の点から劣化がほとんどなく且つ演算量が14-46%削減されることが分かった．まず，整数画素精度の動き推定マッチング誤差を，縦横画素位置を変数とする2次元データとして取扱い誤差曲面を生成した．次に，その曲面に対する数学的モデルとして，9個あるいは6個のパラメタからなる放物線関数(パラボリック形式)を用いた．予備実験を行い，実験から得られた誤差曲面とモデル関数とを照合した．その結果，曲面のある小領域では，中心位置の誤差演算時において45度方向に位置する画素の誤差からの寄与が非常に少ないことが分かった．したがって，中心位置とその縦横４画素からなる誤差に与える５個の係数（パラメタ）からなるモデルにより，十分誤差を近似できることが分かった．さらにより簡単な縦横分離形式でも十分な精度で近似できることが分かった．この１次元放物線を決定するパラメタの定義域についても検討した．また，放物線と比べてより自由度の高いいくつかのスプライン曲線の導入について検討した．スプライン曲線には，Cardinal-Spline, Hermite-Spline, Bézier-Spline, Non-Uniform Rational B-Spline (NURBS) などがある．このうち，3点の制御点で描ける2次形式のBézier-Splineによる誤差表面近似に基づく少数画素精度動き推定手法を提案した．実験の結果，整数画素精度の動き推定にBézier-Splineに基づく提案手法を組み合わせた方式は，より精度の細かなベクトルを選ぶ傾向にあることがわかった．Bézier-Splineの曲線を規定する係数の導出において、しきい値により場合分けすると，より性能が向上することが分かった．そこで提案方式をMPEG-HEVCと組み合わせ，その性能評価を行った．HEVCはAVCと比較して，符号化画像が同一の品質であれば，符号化レートが50%に削減できる超高圧縮符号化方式であり，2013年1月に仕様が凍結された最新の方式である．AVCは現在では幅広く普及しており，Blu-Ray Diskやハンディカム用のAVCHDで用いられている高圧縮符号化方式である．AVCでは動き補償のためのブロックサイズは主に16x16画素であったが，HEVCでは動き補償のためのブロックサイズが64x64画素にまで拡大可能となっている．したがってBézier-Splineの曲線を決定するしきい値を調節することとした．AVC，HEVCいずれの方式と組み合わせても，提案方式はR-Dの観点から劣化はなく，演算量が14-46%に低減できることが分かった．また処理時間をCPU time (msec) で評価した場合でも，13-36%に低減できることが分かった．したがって本提案方式は，8k画像 (水平画素数8000) のように超高精細に向かいつつある今後のテレビジョン画像を対象とする符号化器や，モバイル端末のようにCPU能力が低い符号化器において，演算処理量の低減に大きく貢献する技術である．