Recognition of object contours from stereo images: an edge combination approach

Abstract

We present an algorithm to combine edge information from stereo-derived depth maps with edges from the original intensity/color image to improve the contour detection in images of natural scenes. After computing the disparity map, we generate a so-called "edge combination image", which relies on those edges of the original image that are also present in the stereo map. We describe an algorithm to identify corresponding intensity and depth edges, which are usually slightly displaced due to nonperfect stereo reconstruction. Our experiments demonstrate how the proposed edge combination approach can be used in conjunction with an active contours algorithm to achieve better segmentation results.