A New Statistically Adaptive Block-Matching Criterion for Motion Estimation
Seung-Hwan Kim, Hong Kook Kim, Yo‐Sung Ho · ITC-CSCC :International Technical Conference on Circuits Systems, Computers and Communications · 2004
In this paper, we address the issue of motion estimation in real video sequences. While the conven- tional block-matching criteria for motion estimation is based on the assumption that the average values of the target block and reference blocks are identical. This causes performance degradation for motion estimation because average values are time-varying in real video sequences. In order to alleviate this problem, we propose a new statistically adaptive block-matching criterion (SABMC). In SABMC, the motion vector is estimated by considering time-varying average and shape that repre- sents the fluctuation of pixel values separately, because they are of different importance. Simulation results dem- onstrate that the proposed SABMC criterion provides higher PSNR values than conventional block-matching criteria at various bit rates.