Disparity estimation with edge-based matching and interpolation

Stereo matching is one of the most active research areas in computer vision. In this paper, we present a disparity estimation algorithm using edge matching and spatial interpolation algorithm on color information. Firstly, we select some one to one corresponding points as seed points, then propagate them to edge map. Color based Gaussian weighted spatial interpolation is utilized on edge disparity map to obtain the final disparity map, which partly resolves the problem of occlusion. The performance of the algorithm is illustrated by simulations carried out on some pairs of images.