Redundant-Key-Picture Error Resilient Coding with Motion-Copy Error Concealment in Scalable Video Coding

摘要

Error resilient coding and error concealment techniques have been introduced in scalable video coding to reduce the quality degradation impact of transmission errors. Redundant picture is one of the error resilient tools, appropriate for delay sensitive applications where retransmission is not possible and channel loss rates such as in packet-based wireless networks, are not a priori known. In this paper, combination of redundant-key-picture error resilient coding with motion-copy error concealment at the decoder is proposed based on experimental evaluation. Experimental results demonstrate that significant PSNR gain can be achieved using key picture error concealment with motion-copy for various tested sequences under different packet loss rates compared to JVT-W049.