Disparity estimation using edge preserving method

A disparity map is usually obtained by stereo images using a stereo matching method. Edge boundaries in the disparity map separate two different objects. Therefore, edge preservation is one of important issues in the stereo matching method. A conventional distance transform was proposed to preserve edge boundaries in the disparity map. However, this method has a complexity problem because of its iterative calculation. To reduce the calculation complexity, we propose a modified distance transform. Another important issue in the stereo matching method is an occlusion handling. We improve the quality of occlusion regions using disparity values of neighboring pixels. We apply the occlusion handling method to the edge preserved disparity map. Experiment results show that the proposed method is faster than the conventional method with a small loss in bad pixel rates.