Zero-synthesis view difference aware view synthesis optimization for HEVC based 3D video compression

Siwei Ma, Shiqi Wang, Wen Gao

Abstract: In this paper, we explore a new scheme for view synthesis optimization (VSO) in the upcoming high efficiency video coding (HEVC) based 3D video compression standard. With the dynamic range analysis of depth value, we present a novel zero-synthesis view difference (ZSVD) model. This model accounts for the compound impact of depth-disparity mapping, texture masking and occlusion information. We further incorporate the ZSVD profile into rate distortion optimization process via the ZSVD adaptive VSO, which is developed by pruning the conventional view rendering algorithm. Experimental results exhibit that the proposed scheme can remarkably reduce the computational complexity with negligible performance loss.