论文标题
基于几何的闭塞意识无监督的立体声匹配以自动驾驶
Geometry-based Occlusion-Aware Unsupervised Stereo Matching for Autonomous Driving
论文作者
论文摘要
最近，基于无监督学习的自动驾驶中有许多立体声匹配方法。他们中的大多数利用重建损失来消除对差异地面损失的依赖。闭塞处理是立体声匹配中的一个具有挑战性的问题，尤其是对于无监督的方法。以前的无监督方法未能充分利用遮挡处理中的几何特性。在本文中，我们介绍了一种检测遮挡区域的有效方法，并提出了一种新颖的无监督训练策略，以处理仅使用预测的左差图图，以迭代方式使用其几何特征。在训练过程中，我们将预测的左侧差图视为伪地面图，并使用几何特征进行遮挡区域。然后，将产生的遮挡面膜用于训练，后处理或两者作为指导。实验表明，我们的方法可以有效地处理遮挡问题，并显着优于其他无监督的立体匹配方法。此外，我们的闭塞策略可以方便地扩展到其他立体声方法并改善其性能。
Recently, there are emerging many stereo matching methods for autonomous driving based on unsupervised learning. Most of them take advantage of reconstruction losses to remove dependency on disparity groundtruth. Occlusion handling is a challenging problem in stereo matching, especially for unsupervised methods. Previous unsupervised methods failed to take full advantage of geometry properties in occlusion handling. In this paper, we introduce an effective way to detect occlusion regions and propose a novel unsupervised training strategy to deal with occlusion that only uses the predicted left disparity map, by making use of its geometry features in an iterative way. In the training process, we regard the predicted left disparity map as pseudo groundtruth and infer occluded regions using geometry features. The resulting occlusion mask is then used in either training, post-processing, or both of them as guidance. Experiments show that our method could deal with the occlusion problem effectively and significantly outperforms the other unsupervised methods for stereo matching. Moreover, our occlusion-aware strategies can be extended to the other stereo methods conveniently and improve their performances.
