Performance enhancement for rate-constrained motion estimation using Kalman filter
Chung‐Ming Kuo, Shu-Chiang Chung, Po-Yi Shih · Optical Engineering · 2004
We present a new rate-constrained [rate and distortion (R-D)] motion estimation algorithm using Kalman filter to enhance the performance of the conventional R-D motion estimation at a relative low computational cost. We try to amend the incorrect and/or inaccurate estimate of motion with higher precision by using a Kalman filter. We first obtain a measurement of the motion vector of a block by using the existing R-D motion estimation scheme. We then generate the predicted motion vector utilizing the interblock correlation in neighboring blocks. Based on these two motion information, a simple 1-D Kalman filter is applied to obtain the optimal estimate of the motion vector. Simulation results show that the proposed technique can efficiently improve the R-D motion estimation performance.