Yan Tingman, Gan Yangzhou, Xia Zeyang, Zhao Qunfei
IEEE Trans Image Process. 2019 Aug;28(8):3885-3897. doi: 10.1109/TIP.2019.2903318. Epub 2019 Mar 6.
In this paper, we propose a disparity refinement method that directly refines the winner-take-all (WTA) disparity map by exploring its statistical significance. According to the primary steps of the segment-based stereo matching, the reference image is over-segmented into superpixels and a disparity plane is fitted for each superpixel by an improved random sample consensus (RANSAC). We design a two-layer optimization to refine the disparity plane. In the global optimization, mean disparities of superpixels are estimated by Markov random field (MRF) inference, and then, a 3D neighborhood system is derived from the mean disparities for occlusion handling. In the local optimization, a probability model exploiting Bayesian inference and Bayesian prediction is adopted and achieves second-order smoothness implicitly among 3D neighbors. The two-layer optimization is a pure disparity refinement method because no correlation information between stereo image pairs is demanded during the refinement. Experimental results on the Middlebury and KITTI datasets demonstrate that the proposed method can perform accurate stereo matching with a faster speed and handle the occlusion effectively. It can be indicated that the "matching cost computation + disparity refinement" framework is a possible solution to produce accurate disparity map at low computational cost.
在本文中，我们提出了一种视差细化方法，该方法通过探索胜者全得（WTA）视差图的统计意义来直接对其进行细化。根据基于分割的立体匹配的主要步骤，将参考图像过度分割为超像素，并通过改进的随机抽样一致性（RANSAC）为每个超像素拟合一个视差平面。我们设计了一个两层优化来细化视差平面。在全局优化中，通过马尔可夫随机场（MRF）推理估计超像素的平均视差，然后，从平均视差导出一个3D邻域系统用于遮挡处理。在局部优化中，采用了一种利用贝叶斯推理和贝叶斯预测的概率模型，并在3D邻域之间隐式地实现了二阶平滑。两层优化是一种纯粹的视差细化方法，因为在细化过程中不需要立体图像对之间的相关信息。在Middlebury和KITTI数据集上的实验结果表明，所提出的方法能够以更快的速度进行精确的立体匹配，并有效地处理遮挡。可以表明，“匹配成本计算+视差细化”框架是一种以低计算成本生成精确视差图的可能解决方案。
