一种低复杂度立体视频错误隐藏新方法

A new low-complexity error concealment method for stereo video

针对立体视频流传输中右视点整帧丢失,提出了一种低复杂度的错误隐藏算法。首先,为了高效地感知立体视频的时域质量和视点间质量,定义了时域相似尺度(TSM)、视间相似尺度(ISM)的概念;将前一时刻右视点图像进行时域和视点间匹配,分别求取其以像素为单位的TSM和ISM映射图;然后,计算前一时刻右视点图像当前宏块的TSM和ISM值,通过比较得到当前宏块的预测模式;最后,根据视频序列的时域一致性,将前一时刻右视点图像宏块的预测模式作为丢失图像宏块的预测模式,从而使用运动补偿预测(MCP)或者视差补偿预测(DCP)的方法恢复丢失信息。研究结果表明,与已有错误隐藏算法相比,本文算法获得更好主客观视觉效果;同时与基于图像结构相似度(SSIM)的错误隐藏算法相比,在保持主观视觉质量情况下,错误隐藏时间节省20%左右。

In this paper,to deal with the whole-frame loss of the right view in compressed stereo video stream transmission,a low-complexity error concealment algorithm is proposed.Firstly,in order to perceive the temporal and inter-view quality of the stereo video,a fast strategy is proposed based on the concepts of temporal similarity metric（TSM）and inter-view similarity metric（ISM）.Then,macroblocks（MBs）in the previous frame of the lost frame are matched in both temporal and inter-view domains to obtain the pixel-based TSM and ISM maps.For each MB in the lost right-view frame,its TSM and ISM values are calculated to obtain the MB prediction mode（MPM）.Finally,according to the temporal consistency of video sequences,MPMs of the previous right-view frame are used as the MPM of the lost frame.Either motion compensation or disparity compensation can be quickly decided and used for resilience of the lost right-view frame by the prediction mode.Experimental results show that compared with the existing error concealment algorithms,the proposed algorithm has superior subjective and objective qualities,and compared with the error concealment algorithm based on structural similarity（SSIM）,its error concealment time is reduced by about 20% with almost the same perceptual quality.