Optimized video coding for omnidirectional videos

PubDate: August 2017

Teams: Tsinghua University

Writers: Minhao Tang ; Yu Zhang ; Jiangtao Wen ; Shiqiang Yang

PDF: Optimized video coding for omnidirectional videos

Abstract

The ever widening application of virtual reality requires the ultra high resolution omnidirectional videos (OVs) to be transmitted over the wired and wireless Internet at low cost (i.e. bitrate). Various solutions have been proposed to intelligently reduce the bitrate, e.g. adapting the spatial resolution of the video for different directions of the panorama with regard to current direction that the viewer is looking at and the distribution of the probability of each direction to be watched according to the video content. Due to various reasons, spatial resolution adaptation may often cause perceivable quality degradation to user experience. In this paper, we proposed two adaptive encoding techniques to reduce the bitrate of OVs after compression. The first is a content adaptive temporal resolution adaptation scheme for OVs using cube map projection. The second is a quantization and rate-distortion optimization scheme for equirectangular projection. Experiments implementing the proposed algorithms in the open source HEVC encoder x265 show that the proposed algorithms can save 14.69% and 13.01% in BD-rate on average for the cube map and equirectangular projected OVs respectively, while no degradation to the visual experience was reported in subjective tests. The proposed algorithms are compatible with and therefore can collaborate with the currently used adaptive spatial resolution scheme for further bitrate reduction.