摘要
针对高效视频编码中运动估计的计算复杂度高和消耗大量编码时间的问题,提出一种改进的TZS算法,从而有效提升视频编码效率.首先对搜索区域进行运动矢量预测,确定起始搜索点;再使用宽钻石搜索算法代替钻石搜索算法来搜索率失真最小点;然后根据运动矢量的十字中心偏置的分布特点,采用八边形-十字栅格模式进行栅格搜索,减少编码过程中的搜索点数;最后采用精细搜索得到最佳匹配点.实验结果表明,提出的算法与HM-16.14中的TZS标准算法相比,平均降低了26.59%的编码时间,并保持视频质量基本不变.
Considering the high computational complexity and high coding time of motion estimation in High Efficiency Video Coding,a modified TZS algorithm is proposed to effectively improve the efficiency of video coding.Firstly,the search area is predicted by motion vector to determine the actual search point.Then,WDS search algorithm is used instead of DS search algorithm to search the minimum distortion point.Next,according to the distribution characteristics of the cross center bias of the motion vector,the Octagonal-Axis Raster Pattern is used for raster search to reduce the number of search points in the coding process.Compared with TZS standard algorithm in HM-16.14,experimental results show that the proposed algorithm reduces 26.59%coding time on average and keeps the similar video quality.
作者
但鸿键
汪伟
DAN Hong-jian;WANG Wei(School of Optical-Electric and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2021年第7期1421-1426,共6页
Journal of Chinese Computer Systems
基金
上海市教委2016年青年教师培养计划项目(ZZs115012)资助
上海理工大学光电学院教师创新能力建设项目(1000302006)资助。