Stereo Matching Based on Edge-Aware T-MST

摘要

In this paper, a novel dense stereo matching algorithm is proposed based on edge-aware truncated minimum spanning tree (T-MST). Instead of employing non-local cost aggregation on traditional MST which is only generated from color difference of neighbouring pixels, a new tree structure, "Edge-Aware T-MST", is proposed to aggregate the cost according to the image texture. Specifically, cost aggregation is promoted among highly textured region. Meanwhile, the "edge fatten" effect is restrained by combining edge-prior and superpixel-prior together to locate the true disparity edge. Then a widely used Winner-Take-All (WTA) strategy is performed to establish initial disparity map. An adaptive non-local refinement is also performed based on the stability of initial stereo estimation. Experimental results on Middlebury data set show that our algorithm outperforms the current state-of-the-art non-local MST-based stereo matching algorithms.