Low complexity scalable image and video coding for Internet applications.

摘要

The demand on image and video bistream features, like SNR and resolution scalability, is increasing because of the increasing need of efficient data manipulation. Complexity issue is also of major concern in algorithm design. In this research, we focus on developing low complexity scalable image and video coding algorithms and investigating their applications in Internet environment.; We develop a high performance block-based wavelet image coder of very low implementational complexity yet of rich features. In this image coder, the Dual-Sliding Wavelet Transform (DSWT) is first applied to image data to generate wavelet coefficients in fixed-size blocks. Coefficient blocks are directly coded with the Low Complexity Binary Description (LCBiD) coding algorithm. There is also no intermediate buffering needed in between DSWT and LCBiD. Compressed bistream generated is both SNR and resolution scalable, as well as highly resilient to transmission errors. This codec has a very good conditioning performance even the block size (16, 16).; Based on this image codec, a new framework for scalable video coding is developed. One major difficulty in designing a scalable video codec is how to obtain a highly scalable video stream while the inter-frame dependency is still efficiently exploited. Layered Motion Compensation and Coding is our solution to this problem. This video coder encodes some frames as I-frames and others as P-frames. An I-frame is encoded by an LCBiD-like image coder. Coefficient domain motion compensation is performed in encoding a P-frame. The prediction residue is calculated in a way that the layered structure is strictly enforced across the entire Group of Pictures. The residue coding scheme is also based on LCBiD with the introduction of conditional sign-bit coding scheme to achieve an SNR scalable video stream. The resolution scalability is realized by coding each subband independently.; An RTSP-based media streaming framework has been constructed to deliver the scalable video stream. The SNR scalability of fine granularity enables the delivery of the video streams of different qualities by performing simple parsing operations.