One-View Occlusion Detection for Stereo Matching with a Fully Connected CRF Model

© 1992-2012 IEEE. In this paper, we extend the standard belief propagation (BP) sequential technique proposed in the tree-reweighted sequential method to the fully connected conditional random field models with the geodesic distance affinity. The proposed method has been applied to the stereo matching problem. Also, a new approach to the BP marginal solution is proposed that we call one-view occlusion detection (OVOD). In contrast to the standard winner takes all estimation, the proposed OVOD solution allows finding occluded regions in the disparity map and simultaneously improve the matching result. As a result, we can perform only one energy minimization process and avoid the cost calculation for the second view and the left-right check procedure. We show that the OVOD approach considerably improves the results for cost augmentation and energy minimization techniques in comparison with the standard one-view affinity space implementation. We apply our method to the Middlebury data set and reach the state-of-the-art, especially for median, average, and mean squared error metrics.