Advances in Multimedia Information Processing, PCM 2012: 13th Pacific-Rim Confe [Paperback]

This book constitutes the proceedings of the 13th Pacific Rim Conference on Multimedia, held in Singapore during December 4-6, 2012.The 59 revised full papers presented were carefully reviewed and selected from 106 submissions for the main conference and are accompanied by 23 presentations of 4 special sessions. The papers are organized in topical sections on multimedia content analysis, image and video processing, video coding and multimedia information processing, image/video processing and analysis, video coding and multimedia system, advanced image and video coding, cross media learning with structural priors, as well as efficient multimedia analysis and utilization.Multimedia Content Analysis I Incremental Learning of Patch-Based Bag of Facial Words Representation for Online Face Recognition in Videos.- Evaluation of Audio Quality Requirements over Extended Periods of Time Using Long Duration Audiovisual Content.- Hashing with Cauchy Graph.-Multimedia Event Detection Using Segment- Based Approach for Motion Feature.- Robust Feature Bundling.- Image and Video Processing I.- Colorization for Gray Scale Facial Image by Locality-Constrained Linear Coding.- New Eye Contact Correction Using Radial Basis Function for Wide Baseline Videoconference System.- An Integrated Hole-Filling Algorithm for View Synthesis.- A Real-Time On-Chip Algorithm for IMU-Based Gait Measurement.- Reducing the Power Consumption of an IMU- Based Gait Measurement System.- Video Coding and Multimedia Information Processing ISSIM-Based End-to-End Distortion Modeling for H.264 Video Coding.- A Videoconferencing-Oriented Hybrid-Domain H.264/SVC to H.264/AVC Spatial Transcoder.- Robust Noise Estimation Based on Noise Injection.-Image Quality Assessment Based on Improved Structural SIMilarity.- Multi-hypothesis Temporal Prediction Using Template MatchingPrediction and Block Motion Compensation for High Efficiency Video Coding.-Multimedia Content Analysis II Leveraging Exemplar and Saliency Model forló‡