Yan. L, Gauthier. L.
An occlusion-aware framework is proposed to robustly estimate the disparities of light field images. It is mainly realized by leveraging multiple edge cues to occlusion detection and then integrate it with local costs into an energy function. To check the performance, the quantitative and/or qualitative evaluations are performed on both synthetic and natural light field datasets. It demonstrates that the proposed framework is robust to the density and disparity range of the light field, advancing the state-of-the-art light field disparity estimation frameworks on aspect of accuracies.