A multi-window stereo matching algorithm in rank tranform domain

A novel stereo video matching algorithm based on multiple searching windows in rank transform domain is proposed in this paper. By doing a rank transform, stereo videos are mapped to the rank space from the gray-level space, which eliminate the interference of the environment variations and diverse cameras. The disparity is estimated by a multi-window algorithm with fast searching strategy, which improves the accuracy of the depth map and the matching speed as well. After that, a disparity compensation step is done by smooth area based algorithm to get a smoother depth map and clearer region of objects. The experiment results show the efficiency and the robustness of the proposed method.