基于H.264的多参考帧运动估计快速算法

Fast Multi-frame Motion Estimation Algorithms for H.264

H.264编码器支持7种块类型在多个参考帧中搜索最佳运动矢量。参考代码采用逐一搜索的方式,极大地增加了计算复杂度。该文根据相邻帧间运动矢量和像素的相关性,将菱形算法扩展到多参考帧情形,提出了一种新的块匹配运动估计算法——三维菱形法。实验表明,该算法在保持高信噪比和低比特率的同时,能有效地降低编码器复杂度。

H.264 codec allows seven block sizes to perform the motion compensation, and refers to several frames for searching the best motion vector to achieve lower bitrate and higher quality. Since the reference software uses the full search scheme to obtain the best performance, it will spend a lot of searching time. The motion vectors in near blocks and pints in same blocks between sequent frames are correlated. Considering this characteristics, this paper proposes efficient searching algorithms by extending famous DS algorithm to multiple reference frames. Experimental results show that such method can reduce the complexity considerably while maintaining similar PSNR and bit rate.