Advances in Multimedia Information Processing - PCM 2007

1. Front Matter

   PDF

2. Session-1: Image Classification and Retrieval

   1. FADA: An Efficient Dimension Reduction Scheme for Image Classification

      • Yijuan Lu, Jingsheng Ma, Qi Tian

      Pages 1-9

   2. Using Camera Settings Templates (“Scene Modes”) for Image Scene Classification of Photographs Taken on Manual/Expert Settings

      • William Ku, Mohan S. Kankanhalli, Joo-Hwee Lim

      Pages 10-17

   3. Modeling User Feedback Using a Hierarchical Graphical Model for Interactive Image Retrieval

      • Jian Guan, Guoping Qiu

      Pages 18-29

   4. Graph Cuts in Content-Based Image Classification and Retrieval with Relevance Feedback

      • Ning Zhang, Ling Guan

      Pages 30-39

   5. A Novel Active Learning Approach for SVM Based Web Image Retrieval

      • Jin Yuan, Xiangdong Zhou, Hongtao Xu, Mei Wang, Wei Wang

      Pages 40-49

3. Special Session-1: The AVS China National Standard - Technology, Applications and Products

   1. An End-to-End Application System of AVS: AVS-IPTV in China

      • Wen Gao, Siwei Ma, Cliff Reader

      Pages 50-57

   2. An AVS Based IPTV System Network Trial

      • Zhifeng Jiang, Wen Gao, Hongqi Liu

      Pages 58-64

   3. Usage of MPEG-2 to AVS Transcoder in IPTV System

      • GuoZhong Wang, HaiWu Zhao, Guowei Teng

      Pages 65-70

   4. The Audio and Video Synchronization Method Used in AVS System

      • Zhijie Yang, Xuemin Chen, Jiang Fu, Brian Heng

      Pages 71-77

4. Session-2: Human Face and Action Recognition

   1. Expression-Invariant Face Recognition with Accurate Optical Flow

      • Chao-Kuei Hsieh, Shang-Hong Lai, Yung-Chang Chen

      Pages 78-87

   2. Real-Time Facial Feature Point Extraction

      • Ce Zhan, Wanqing Li, Philip Ogunbona, Farzad Safaei

      Pages 88-97

   3. Local Dual Closed Loop Model Based Bayesian Face Tracking

      • Dan Yao, Hong Lu, Xiangyang Xue, Zhongyi Zhou

      Pages 98-107

   4. View-Independent Human Action Recognition by Action Hypersphere in Nonlinear Subspace

      • Jian Zhang, Yueting Zhuang

      Pages 108-117

   5. Efficient Adaptive Background Subtraction Based on Multi-resolution Background Modelling and Updating

      • Ruijiang Luo, Liyuan Li, Irene Yu-Hua Gu

      Pages 118-127

5. Session-3: H.264 Video Coding

   1. Towards a Stringent Bit-Rate Conformance for Frame-Layer Rate Control in H.264/AVC

      • Evan Tan, Jing Chen

      Pages 128-137

   2. A Quantized Transform-Domain Motion Estimation Technique for H.264 Secondary SP-Frames

      • Ki-Kit Lai, Yui-Lam Chan, Wan-Chi Siu

      Pages 138-147

   3. Efficient Intra Mode Decision Via Statistical Learning

      • Chiuan Hwang, Shang-Hong Lai

      Pages 148-157

   4. Fast Mode Decision Algorithms for Inter/Intra Prediction in H.264 Video Coding

      • Ling-Jiao Pan, Seung-Hwan Kim, Yo-Sung Ho

      Pages 158-167

   5. A Novel Fast Motion Estimation Algorithm Based on SSIM for H.264 Video Coding

      • Chun-ling Yang, Hua-xing Wang, Lai-Man Po

      Pages 168-176

The Pacific-Rim Conference on Multimedia (PCM) was held in Hong Kong, at the City University of Hong Kong, during December 11–14, 2007. Started in 2000, PCM has been held in various places around the Pacific Rim, including Sydney, Beijing, Hsinchu, Singapore, Tokyo, Jeju, and Zhejiang in chronological order. PCM is a major annual international conference organized as a forum for the dissemination of state-of-the-art technological advances and research results in the fields of theoretical, experimental, and applied multimedia analysis and processing. PCM 2007 was organized into 5 different tracks with a total of 247 submissions from 26 countries and regions including Australia, Belgium, Canada, China, Japan, Korea, New Zealand, Singapore, Spain, Taiwan, the UK and the USA. Among the five tracks, “multimedia analysis and retrieval” received the most number of submissions (34% of the submissions). After a rigorous review process, 73 papers were accepted for oral presentations, giving an acceptance rate of under 29% for PCM 2007. In addition, 21 papers were accepted for poster presentations. We would like to thank all the Track Chairs and the reviewers for their timely handling of the paper reviews. We are particularly grateful to Chong-Wah Ngo and his team for their support of the Web-based review system throughout the process. We are also indebted to the Special Sessions Chairs, Qi Tian and Timothy Shih, for the organization of the two special sessions on “The AVS China National Standard” and “Multimedia Information System for Biomedical Research,” respectively.

• 3D
• 3D video
• DRM
• H.264
• Multimedia
• Session
• Textur
• architecture
• data compression
• digital signature
• face recogition
• image retrieval
• motion analysis
• rendering
• security

Policies and ethics