A Fast Motion Estimation Algorithm Using Dynamically Adaptive Rood Pattern search Algorithm

摘要

Motion Estimation (ME) for video coding has taken an entirely different direction with the emergence of Adaptive Rood Pattern (ARPS) algorithm. ARPS significantly reduce the tremendous computational overhead of motion estimation that was incurred in a video encoding system. Based on ARPS, a new Adaptive Rood Pattern search Algorithm which utilizes the Spatial correlations of neighboring blocks (hence denoted it by ARPS_S) for the Motion Vector (MV) prediction is proposed in this paper. Unlike ARPS, in ARPS S the initial center point is positioned at the point of the predicted MV, and then a more flexible strategy is devised which adjusts the lengths of the horizontal and vertical rood arms adaptively and separately. After the initial search, a fixed-size rood pattern is employed repeatedly until the best matching block is found. Experimental results show that ARPSS out performs ARPS algorithms in terms of the search speed, PSNR and Mean Absolute Difference (MAD).