精确像素引导的匹配
Pixel-Accurate Epipolar Guided Matching
AI总结：
本文提出一种精确引导匹配方法，通过在角度空间中直接选择候选，提高鲁棒性和效率，实验表明在ETH3D数据集上速度提升显著。
摘要：
关键点匹配在重复纹理或大基线视图等挑战性条件下往往速度慢且不可靠。在这种情况下，已知的几何关系（例如基础矩阵）可以用来将潜在对应关系限制在狭窄的视角 envelope 中，从而减少搜索空间并提高鲁棒性。这些视角引导匹配方法在如SfM等任务中已证明有效；然而，大多数方法依赖于粗略的空间分箱，这会引入近似误差，需要昂贵的后期处理，并可能导致遗漏有效对应关系。我们通过一种精确的公式来解决这些限制，该公式在视角空间中直接进行候选选择。在我们的方法中，每个关键点被分配一个容忍圆，当从视点观看时，定义一个角度区间。匹配则变成一个1D角度区间查询，通过段树高效解决，耗时对数时间。这保证了像素级的容忍度，支持每个关键点的控制，并消除了不必要的描述符比较。在ETH3D上的广泛评估表明，与现有方法相比，速度显著提升，同时恢复了精确的对应集。
英文摘要：
Keypoint matching can be slow and unreliable in challenging conditions such as repetitive textures or wide-baseline views. In such cases, known geometric relations (e.g., the fundamental matrix) can be used to restrict potential correspondences to a narrow epipolar envelope, thereby reducing the search space and improving robustness. These epipolar-guided matching approaches have proved effective in tasks such as SfM; however, most rely on coarse spatial binning, which introduces approximation errors, requires costly post-processing, and may miss valid correspondences. We address these limitations with an exact formulation that performs candidate selection directly in angular space. In our approach, each keypoint is assigned a tolerance circle which, when viewed from the epipole, defines an angular interval. Matching then becomes a 1D angular interval query, solved efficiently in logarithmic time with a segment tree. This guarantees pixel-level tolerance, supports per-keypoint control, and removes unnecessary descriptor comparisons. Extensive evaluation on ETH3D demonstrates noticeable speedups over existing approaches while recovering exact correspondence sets.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
