| Along the improvements of network bandwidth and processor capability, it is noted that the multimedia communication is coming to be the key component of next-generation communication system. Since the video communication is the inevitable part of the multimedia communication system, video coding is more and more important and necessary. Co-sponsored under the National High Technology Research and Development Program of China (863 Program)– Joint Source-Channel Coding (Grant No: 2002AA119010), National Science Fund of China– Research on general Rate-Distortion Model on Wireless Channel (Grant No.: 60372067)and Youth Chenguang Program of Wuhan– Reseach on Joint Power-Rate-Distortion Model for Wireless Multimedia Communication (Grant No.: 20045006071-18), we have participate into the AVS (Audio Video Coding Standard Workgroup of China) workgroup actively including proposing the technical proposals, chairing ad-hoc core experiment group and pushing the standard advancing domestically. Together with the advance of wireless technologies, AVS-M is standardized under the efforts from AVS experts for the mobile application with high coding efficiency and low complexity.In this thesis, at first, the innovative techniques adopted in AVS-M are reviewed specifically. Based on the AVS-M platform, the complexity metric is proposed which is ultilized to analyze the signal processing complexity of AVS-M codec. Note that the complexity metric is also suitable for another video coding standard based video coding analysis, complexity estimation and conformace codec developments instruction.Based on the complexity metrics and video coding fundaments, the most time and complexity consuming part– motion estimation in encoder pair is discussed and improved with complexity reduction. For the decoder part, the macroblock-based decoder (AVS-M Golden C decoder) is first designed for resource-limited applications. Based on the AVS-M Golden C decoder platform, the specific complexity analysis is illustrated by the table description algorithm and the complexity reduction proposal is presented to help the decoder optimization. For the resource limited wireless application environment, since the high processing complexity and volumeful data transferring, the processing complexity of AVS-M codec is analyzed and detailed in this thesis. Note that the proposed complexity metric is intructive for the platform dependent codec realization. Based on the complexity analysis in this thesis, there are some improvements in associated aspects of video processing including the reduction of development cost, optimized complexity control, etc. |
PDF Full Text Request