Region-based Horn-Schunck optical flow estimation

Abstract

In this paper we propose an efficient modification for the Horn-Schunck optical flow estimation algorithm. The proposed modification is represented by incorporating the segmentation with the optical flow estimation in two-stage optical flow estimation. In the first stage, a reference image is segmented into homogeneous regions. In the second stage, the optical flow is estimated for each region rather than pixels or blocks. In this modification, all pixels in each homogeneous region are assumed to have the same motion vector. The modified Horn-Schunck algorithm takes advantage of image segmentation to overcome problems of conventional optical flow estimation algorithms, which are the handling of un-textured regions, the estimation of correct flow vectors near motion discontinuities and the susceptibility to noise. To demonstrate the efficiency of the proposed algorithms, it is compared with three state-of-the art algorithms including the conventional Horn-Schunck algorithm. Based on the simulation results, the proposed modification has a great effect on the performance of the Horn-Schunck algorithm.