摘要
为增强视频传输的稳健性和提高视频传输的质量,提出了一种基于H.264的多描述视频编码方案以及整帧丢失时的误码掩盖方案。该方案在编码器端编码之前进行必要的预处理工作,利用标准的H.264编码器产生多描述码流及相关的编码信息,在解码器端解码之后进行必要的后处理,再根据接收到的编码信息进行多路视频的加权合成以得到较好的视频质量。整帧丢失时的误码掩盖方案利用其他描述正确解码的相应帧进行替换的方法,误码掩盖之后进行视频加权合成。实验结果证明该方案在各种条件下均取得了相当大的增益,特别是当2路视频质量相差较大时,增益可以达到2.38 dB,在发生整帧丢失的时,也有几个dB的增益。
The robustness and quality of video transmissions by a multiple description video coding algorithm based on the H. 264 standard with an error concealing scheme when whole frames are lost. The algorithm uses the standard video encoder with a pre-processing stage to generate multiple descriptions and the relative coding information. The decoder then uses a post-processing algorithm and side information to combine the descriptions to provide better image quality. The error concealing scheme uses the corresponding frame from another description of the video to replace the lost frame. Test results show that the algorithm significantly improves video quality, especially when large quality differences occur between the two descriptions, with gains of up to 2.38 dB. When wholes frames are lost, the gain will be several dB.
出处
《清华大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2007年第10期1622-1626,共5页
Journal of Tsinghua University(Science and Technology)
基金
国家自然科学基金资助项目(60572081)
关键词
多描述视频编码
视频传输
H.264
误码掩盖
multiple description video coding
video transmission
H. 264
error concealing
