摘要：

提出一种适用于AVS-M基于预测的亚像素运动估计快速搜索算法。分析亚像素候选点的匹配误差关系以排除可能性小的点，保证亚像素搜索的准确性。采用有效的搜索策略，对于不同运动特征的视频序列，在平均PSNR下降不超过0.01 dB的情况下，该算法与亚像素全搜索算法相比，减少了68.50％～87.69％的搜索点数，与中心偏置亚像素搜索算法相比，减少了44.68％～67.38％的搜索点数。

关键词: 视频编码, 亚像素运动估计, 预测矢量

Abstract:

This paper proposes a kind of fast search algorithm which is apt for AVS-M based on predicted sub-pixel Motion Estimation(ME). The algorithm expels the less possibility points by doing matching error relation of sub-pixel points selected analysis, which assures the accuracy of sub-pixel search. Through effective search strategy, compared with the half-pel full search algorithm and Center Biased Fractional Pixel Search(CBFPS), experiments show that the proposed algorithm can respectively reduce the sub-pel search points by 68.50％～87.69％ and 44.68％～67.38％ on average PSNR lost less than 0.01 dB to the video sequences with different motion characteristics.

Key words: video coding, sub-pixel Motion Estimation(ME), prediction vector

中图分类号: 

• TN919.81