An Improved 3d Wavelet-Based Scalable Video Coding Codec for Mc-Ezbc
Ying Chen,Guizhong Liu,Juncai Yao
DOI: https://doi.org/10.1007/s11042-016-3387-1
IF: 2.577
2016-01-01
Multimedia Tools and Applications
Abstract:With the rapid growth of modern multimedia applications, 3D wavelet-based scalable video coding (SVC) codec has received considerable attention lately because of its high coding performance and flexibility in bitstream scalability. It combines the motion-compensated temporal filtering (MCTF) together with the spatial decomposition to produce an embedded bitstream offering various levels of video quality over the heterogeneous networks. However, in the existing 3D wavelet-based SVC schemes, where the block types for block matching algorithms are limited, weighting matrices for block-wise motion compensation are fixed, and variations in activities of temporal subbands are not considered in the selection of the Lagrange multiplier for mode decision. In this paper, our major contribution is to provide some recent extensions to the well-known scalable subband/wavelet video codec Motion-Compensated Embedded Zero Block Coding (MC-EZBC) using three novel and content adaptive algorithms. Firstly, the enhanced hierarchical variable size block matching (Enhanced HVSBM) algorithm is proposed for the variable block size motion estimation. Then, the rate-distortion optimization (RDO) based adaptive Lagrange multiplier selection model for mode decision is presented. Finally, we introduce the adaptive weighting matrices design for overlapped block motion compensation (OBMC). Experimental results show that all the three proposed algorithms significantly improve the overall coding performance of MC-EZBC. Comparisons with other popular wavelet-based SVC codecs demonstrate the effectiveness of our improved codec in terms of both video quality assessment and computational complexity.