Simplified inter-component depth modeling in 3D-HEVC
暂无分享,去创建一个
In this paper, we present a method to reduce complexity of depth modeling modes (DMM), which is currently used in the 3D-HEVC standardization activity. DMM adds four modes to the existing HEVC intra prediction modes; the main purpose is to accurately represent object edges in depth video. Mode 3 of DMM requires distortion calculation of all pre-defined wedgelets. The proposed method employs absolute differences of neighboring pixels in the reference block. The number of wedgelets that need to be concerned can be reduced to six. Experimental results show 3.1% complexity reduction on average while maintaining coding performance, which implies that the correct wedgelet is included, while non-viable wedgelets are disregarded.
[1] Yo-Sung Ho,et al. Efficient multiview depth video coding using depth synthesis prediction , 2011 .
[2] Heiko Schwarz,et al. 3D video coding using advanced prediction, depth modeling, and encoder control methods , 2012, 2012 Picture Coding Symposium.
[3] Aljoscha Smolic,et al. 3D video and free viewpoint video - From capture to display , 2011, Pattern Recognit..