摘要
H.264是由ITU-T最新提出的视频压缩国际标准,同以往视频压缩标准相区别的主要特征就是加大了运动估计和补偿部分的比重,通过改善预测误差来提高编码效率。针对H.264中运动估计和补偿采用的新技术,分析了其为提高编码效率采取的关键措施:可变块大小的运动估计,1/4(甚至1/8)象素精度的运动补偿,多参考帧帧间预测模式等,并通过标准测试序列检验了它的性能,证实其能达到更高的编码效率。
The draft H. 264 standard is a new international video coding standard designed by the ITU - T. A main difference of H. 264 from these existing standard is enlarging the proportion of the motion estimation and compensation to the whole coding, which reduces the prediction residue and provides substantial improvements in coding efficiency. In this paper, we research the new motion estimation and compensation technologies in H. 264. Then we provide a detailed analysis of the coding performance of several key components of the H. 264 motion estimation and compensation model. These components include various combinations of block sizes for motion compensation, fractional pixel motion compensation accuracy and multiple reference frame prediction. Experiments verified its validity using real test sequences and testified that these technologies could improve the coding efficiency.
出处
《淮阴工学院学报》
CAS
2005年第5期17-20,共4页
Journal of Huaiyin Institute of Technology
基金
国家自然科学基金资助项目(60274006)
关键词
H264
运动估计
运动补偿
H. 264
motion estimation
motion compensation
