Semi-global Stereo Matching with Surface Orientation Priors
Semi-global Stereo Matching with Surface Orientation Priors
复制标题
DOI：
10.1109/3dv.2017.00033
复制
发表时间：
2017-09
期刊：
影响因子：
--
通讯作者：
D. Scharstein;Tatsunori Taniai;Sudipta N. Sinha
中科院分区：
文献类型：
--
作者：
D. Scharstein;Tatsunori Taniai;Sudipta N. Sinha
半全局匹配（Semi-Global Matching， SGM）是一种应用广泛的高效立体匹配技术。它适用于有纹理的场景，但由于其正面平行的平滑假设，它不适用于无纹理的倾斜表面。为了解决这个问题，我们提出了一个简单的扩展，称为SGM-P，以利用预先计算的表面取向先验。这种先验倾向于不同2D图像区域或3D场景区域的不同表面倾斜，并且可以通过各种方式推导。在本文中，我们评估了在粗糙分辨率下由立体匹配获得的平面方向先验，并表明这种先验可以在困难的弱纹理场景中产生显着的性能提升。我们还探索了从曼哈顿世界假设中得到的地表正常先验，并使用从地面真实数据中得到的oracle先验分析了潜在的性能增益。SGM- p只给SGM增加了很小的计算开销，是采用高阶平滑项的更复杂方法的一个有吸引力的替代方案。
Semi-Global Matching (SGM) is a widely-used efficient stereo matching technique. It works well for textured scenes, but fails on untextured slanted surfaces due to its fronto-parallel smoothness assumption. To remedy this problem, we propose a simple extension, termed SGM-P, to utilize precomputed surface orientation priors. Such priors favor different surface slants in different 2D image regions or 3D scene regions and can be derived in various ways. In this paper we evaluate plane orientation priors derived from stereo matching at a coarser resolution and show that such priors can yield significant performance gains for difficult weakly-textured scenes. We also explore surface normal priors derived from Manhattan-world assumptions, and we analyze the potential performance gains using oracle priors derived from ground-truth data. SGM-P only adds a minor computational overhead to SGM and is an attractive alternative to more complex methods employing higher-order smoothness terms.