VARIABLE BLOCK-SIZE TRANSFORM AND ENTROPY CODING AT THE ENHANCEMENT LAYER OF FGS (MA-P6)

Author(s) : Jungong Han (ISN national laboratory of Xidian University, China) Xiaoyan Sun (Microsoft Research Asia, China) Feng Wu (Microsoft Research Asia, China) Shipeng Li (Microsoft Research Asia, China) Zhaoyang Lu (ISN national laboratory of Xidian University, China)

Abstract : This paper proposes the variable block-size transform and context-based entropy coding techniques for the enhance-ment layer of FGS (Fine Granularity Scalable) video cod-ing. First, the variable block-size transform is introduced into the enhancement layer to improve the performance of FGS in terms of both visual quality and PSNR. Different from that used in the traditional single layer coding, an R-D selection algorithm is proposed to optimally decide the transform size of each block under consideration of the consistent performance at a range of bit rate. Furthermore, to fully take the advantage of the characteristics and corre-lations of symbols coded in the FGS enhancement layer, different context models are designed for the arithmetic coding according to symbol type and transform size. Ex-perimental results show that the coding efficiency of FGS can be increased 0.2-0.9dB with the proposed techniques.

Menu