A Reduced Orthogonal Search for Video Coding
作者单位：Dept. of ECE Karunya University Dept. of. ECE Karunya University
会议名称：《2011 International Conference on Future Information Technology (ICFIT 2011)》
会议日期：2011年
学科分类：0810[工学-信息与通信工程] 08[工学] 081001[工学-通信与信息系统]
关 键 词：Block Matching Motion Estimation Orthogonal Search Early Termination Video Coding Inner Search
摘 要：In many video coding applications, an efficient fast search motion estimation （ME） algorithm is highly desired. In this paper a new reduced orthogonal search （ROS） algorithm is proposed for fast ME. This algorithm is based on the center-biased characteristics of motion vector to speed up the existing orthogonal search. The performance of this algorithm is evaluated with various video sequences and the results are compared to a traditional well-known full search （FS）, three step search （3SS）, orthogonal search （OS）, and diamond search （DS）. This paper introduces a fast inner-search to early terminate the block matching process in order to reduce computational complexity in the existing OS. Experimental results showed ROS gains less search points of 5 to 11 in average than OS and as compared with previous works, it also outperforms FS, 3SS, DS and OS in most cases with regard to image quality. The advantage of proposed motion estimation technique is further justified by experimental results.