Progressive Description of Video Streaming Media Based on Subjective Perception

Video description is a critical component in video streaming media processing and transmission. Different video data have different levels of contribution to the overall video presentation quality at the receiver side. on the premise of related work, video scene classification, visual sensitive regions(VSR) extraction and MGS(Medium Grain Scalability) encoding, a new approach to progressive description of Video Streaming Media(VSM) based on subjective perception is proposed in this paper. Frames importance according to coding key parameters and slice data importance are determined Firstly. Then VSR and important data protection are jointly considered, a new approach to progressive description of VSM adapted to network bandwidth and quality scalability is proposed. Applying MGS encoding, experimental study is performed on Mobile IP testing platform, the experimental results demonstrate that the proposed approach to progressive description of VSM could effectively protect video important data and significantly improve the video presentation quality.