Long-Rood motion estimation algorithm based on starting search point prediction
Zhaohua Lin, Yanbiao Zou · 2009
Motion estimation algorithm is the most complicated and time consuming part of the video codec. A novel fast block matching motion estimation algorithm named Long-Rood fast motion estimation algorithm based on starting search point prediction is proposed for real-time video coding. To validate the performance, a number of standard test sequences are used. Simulation results demonstrate that the proposed algorithm dramatically outperforms the Exhaustive Search (ES), the Three-Step Search (TSS), the Diamond Search (DS) and the Adaptive Rood Pattern Search (ARPS) algorithms. The simulation results show that the proposed algorithm is about 3.33∼17.12 times faster than that of TSS, 2.12∼8.99 times faster than that of DS, and 1.15∼4.42 times faster than ARPS, while achieving close performance in terms of the Peak Signal-to-Noise Ratio (PSNR).