A Modified Motion Vector Multi-Template Self- Adaptive Search Algorithm for Motion Estimation

摘要：

H.264/AVC has better compression performance because of using motion estimation based on variable block size, but it also increases computational complexity vastly, which restrict the application range of H.264/AVC. The paper draws on ideas of MMVFAST(MuItiMotion Vector Field Adaptive Search Technology), and proposes a modified motion vector multi-template self-adaptive search algorithm of motion estimation, which can reduce the computational complexity of motion estimation effectively. Experimental results show it can retrench time cost of motion estimation and improve speed of encoding greatly on condition that the reconstructed image quality and bit rate keep invariant.

关键词： video coding motion estimation search templates

作者: ZhiGang Zhang CuiLi Ji Qing Liu

作者单位: School of Electronics and Information Engineering, Xian Jiaotong University,Xian, China Informatio Information science Department, Xian University of Technology, Xian, China Engineering Training Center, Xian University of Technology, Xian, China

会议类型: 国际会议

会议名称: 2011 International Conference on Computer Science and Network Technology(2011计算机科学与网络技术国际会议 ICCSNT 2011)

会议地点: 哈尔滨

会议语种:英文

页码: 1525-1528

在线出版日期: 2011-12-24（万方平台首次上网日期，不代表论文的发表时间）