DWT-based Scalable Video Coding --approaches and directions

摘要

Due to the increase demand of multimedia information over the Internet, streaming video has received more and more attention. But the variable of the network-bandwidth and the clients has been threatened this becoming widely used. Scalable video coding is supposed to solving this problem. Unfortunately, the conventional scalable coding methods, including the Fine Granularity Scalability in MPEG-4, are suffering from the drift issue as they are designed over the conventional hybrid coding architecture, thus lead to loss in compression performance. Recently, a new promising technology wavelet-based video coding, which have already demonstrated their potential for quality and resolution scalability in still-image coding as in JPEG2000, seems to discard the drift problem and make the compression performance close to or better than conventional single-layer video coding. There have been some schemes for the compression of video, which most can exploit similar principles to generate embedded bit-streams. In this paper we present most of the existed scalable wavelet-based video-coding technology with some simple compare, and point out what can we do in this topic.