Frame Loss Concealment for Multiview Video Transmission Over Wireless Multimedia Sensor Networks

作者：Yuan Zhou, Wei Xiang, Gengkun Wang · 发表于：IEEE Sensors Journal · 年份：2014 · DOI：10.1109/jsen.2014.2366511 · 被引用次数：47 · 研究领域：Video Coding and Compression Technologies、Image Enhancement Techniques、Advanced Vision and Imaging

In multiview video (MVV) transmission over wireless multimedia sensor networks (WMSNs), a highly compressed video stream is very susceptible to whole-frame losses due to wireless channel errors. Currently, most frame concealment methods can only handle 2-D video or 3-D video with depth, since a lost motion vector can be recovered utilizing the relationship between the color frames and their corresponding depth map. However, depth generation is infeasible in the scenario of WMSNs due to energy constraints and computational limitations of low-power video sensor nodes. In this paper, we tackle the problem of concealing lost whole frames for multiview 3-D video transmission over WMSNs. We propose a new technique, which is capable of dealing with this important problem without the aid of depth map. Experimental results are presented to demonstrate that the proposed algorithm is able to effectively conceal frames of MVV and significantly outperforms other techniques in both objective and subjective quality measurements.