A New Framework of Scalable Video Compression with High Error Resilience Based on Wavelet Using Wyner-Ziv Technique

In recent years, wavelet-based scalable video coding has made a considerable progress, but there are still several problems in the development of WSVC. Such as quantization error accumulation resulted from the open-loop structure of MCTF in coding and the poor error resilient performance of the bit stream. The major goal of this paper is to propose a new framework of scalable video coding based on wavelet using Wyner-Ziv Techniques, which is a new coding structure with high error resilience proposed in resent years. The lowest frequency and the last generated high frequency temporal frames as the base layer are encoded by the EZBC method, and other high frequency temporal frames as the enhancement layer are encoded with Wyner-Ziv coding. The coding efficiency of this scheme is close to MC-EZBC in the error-free scenario and is better than the combination of MC-EZBC and FEC in the error channel.