Full-Text Search:
Home|Journal Papers|About CNKI|User Service|FAQ|Contact Us|中文
《Chinese Journal of Computers》 2006-02
Add to Favorite Get Latest Update

Error Resilience Techniques for Video Communication Based H.264/AVC

ZHOU Ning-Zhao SONG Bin CHANG Yi-Lin (National Key Laboratory of Integrate Services Network, Xidian University, Xi′an 710071)  
H.264/AVC, the result of the collaboration of the ISO/IEC Moving Picture Experts Group and the ITU-T Video Coding Experts Group, is the newest standard for video coding. H.264/AVC can be widely used in many areas because of its higher coding efficiency and the network-friendly video representation. However, Internet is a best-effort network without ensuring QoS especially in real-time service(e.g. video communication), the packets lost will occur in video communication and the quality of the reconstructed image is reduced and even unaccepted. So error resilience techniques are necessary for video communication over Internet. This paper provides some error resilience methods based on H.264/AVC, which include using a forward error correction Tornado code with packet unequal-protection according to different essentiality of video bit-stream, using SEI(Supplemental Enhancement Information)domain to change error corrected ability of Tornado code dynamically to adapt to the environment of the network, and a method of interactive error-propagation prevention, i.e. refreshing intra macroblocks in independent slice between encoder and decoder based SEI domain. Compared with simply using error concealment, the proposed techniques can improve the video quality about 8~15dB over variable and error-prone channel. Furthermore, these methods are compatible with H.264/AVC without any increasing channel burden and arithmetic complexity, and can be used in practice.
【Fund】: 国家自然科学基金(60172030);; 陕西省自然科学基础研究计划项目基金(2004F07);; 华为基金资助~~
【CateGory Index】: TN919.8
Download(CAJ format) Download(PDF format)
CAJViewer7.0 supports all the CNKI file formats; AdobeReader only supports the PDF format.
©2006 Tsinghua Tongfang Knowledge Network Technology Co., Ltd.(Beijing)(TTKN) All rights reserved