用于视频编码的改进的逐次消除算法  

Modified Successive Elimination Algorithm for Video Coding

作　　者：向东[1] 陈加忠[2] 骆正华[3] 

机构地区：[1]武汉理工大学计算机科学与技术学院,武汉430063 [2]华中科技大学计算机科学与技术学院,武汉430074 [3]中南财经政法大学信息学院,武汉430073

出　　处：《武汉理工大学学报》2008年第7期116-118,140,共4页Journal of Wuhan University of Technology

基　　金：湖北省自然科学基金(2005ABA255)

摘　　要：与以往的各种视频编码标准相比,H.264视频编码标准在编码效率上获得了相当大的提高。但这是以提高编码的计算复杂度为代价的,其中运动估计的计算复杂度最大。提出一种适用于全搜索的改进的逐次消除算法,它能够有效地消除运动估计过程中的无效搜索点,加快运动估计的速度。试验结果表明:无效搜索点的消除率最大能够达到98%,而率失真性能无损失。H. 264 has achieved a significant improvement in rate-distortion performance relative to the existing standards. However, this came at a cost in considerably increased complexity at the encoder mainly due to motion estimation. In this paper, a modified successive elimination algorithm applied for full search motion estimation was presented. It could eliminate invalid candidate searching points efficiently and speed the motion estimation procedure. Experimental results showed that the elimination rate of invalid candidate searching points could be up to 98 % without rate distortion performance loss.

关 键 词：视频编码 H.264 运动估计 

分 类 号：TP335[自动化与计算机技术—计算机系统结构]