Stereo Matching Based on Guided Image Filtering with Permeability Filter

The formation of current stereo matching algorithms based on the guided image filtering starts from the minimization of an energy function. We extend this approach by introducing weights to the function such that contribution of different pixels can be assigned individually. This scheme is integrated with the so-called permeability filter for the weighting factors. As the whole picture is involved in the computation, no decision for the window size is required. Extensive experimental comparisons with several state-of-the-art methods using the Middlebury Stereo Evaluation version 3 datasets show that the proposed scheme has a great advantage in stereo matching.