Research Of Delta CPV Decision For Depth Maps In 3D-HEVC

In recent years,3D video has become popular because it can provide users with three-dimensional visual experience.Compared with traditional 2D video,3D video not only contains the texture information of a single viewpoint,but also contains the texture information of other viewpoints and the corresponding depth information.3D-HEVC is the latest 3D video coding standard,on the basis of the traditional prediction technology,it reduces the redundancy among views and improves the coding efficiency with inter-view prediction technology by the full usage of correlation among views.Compared with the texture map,the depth map has completely different applications which is mainly used to synthesize virtual view at the decoder and not to watch directly.So the traditional texture coding method is not suitable for encoding depth map.Therefore,3D-HEVC introduces the Segment-wise DC Coding(SDC)and Depth Modelling Modes(DMMs)to improve the coding quality of depth map.However,the introduction of SDC and DMM greatly increase the complexity of the original algorithm.In order to reduce the time of coding algorithm,this paper mainly studies the CPV decision process of SDC and DMM in 3D-HEVC.1)Optimal Delta CPV fast decision algorithm in intra SDC is set forth in the thesis.In SDC,the optimal Delta CPV is selected from the five candidates values by comparing the RD-cost,which produce great computation cost.The present study improves the existing SDC optimization algorithm,reduces the search ranges and lessens the video coding time.2)Optimal Delta CPV fast decision algorithm in DMMs is put forward in present study.In the intra-prediction process of depth map,after the best partition mode of DMMs is determined,the corresponding Delta CPV is modified.On the one hand,based on the analysis of the Delta CPV's decision process,this paper puts forward the idea that in case the initial Delta CPV is optimal,unnecessary search process is terminated in advance by using the information of the initial value of the search process.On the other hand,the method of using the mean difference to determine the search range is proposed because the mean difference is a good reflection of the average difference between the original signal and the mean value.To sum up,the study proposes two kinds of optimization algorithm for Delta CPV's decision process.It reduces the complexity of depth map coding in the premise of ensuring encoding quality.