基于边缘复杂度的深度图帧内模式选择方法

Edge Complexity Based Intra Mode Selection for Depth-map Coding in 3D-HEVC

3D-HEVC在深度图中的内部模式众多导致了编码时间大量增加.为此,提出了一种基于边缘复杂度的深度图帧内模式选择方法.由于DMMs中的深度图帧内模式适用于不同特征的边缘,其中的Intra_Wedge模式针对简单边缘而Intra_Contour模式适用复杂边缘.因此提出了依据边缘复杂度情况分别选择不同深度图帧内模式到候选列表中.实验结果表明,该算法在保证视频质量的条件下,节省了29.24%的编码时间.

In 3D-HEVC,a large quantity of depth map intra models results in unexpected increase of time consuming.In this paper,a new intra mode decision algorithm by detecting edge complexity is proposed.Depth intra prediction modes in DMMs are applicable for edges of different features,among which Intra Wedge mode is mainly for simple edges and Intral Contour mode is aimed at complex edges.Then based on edge complexity,we add Intra Wedge or Intra Contour to mode candidates instead of all DMMs modes.The experiment results show that our algorithm is capable to save 29.24%of the depth maps coding time without any loss of the quality.