基于在线神经三维回归的稀疏视图定位
Sparse-View Localization via Online Neural 3D Regression
提出在线训练的神经回归器，在无预建3D地图、数据库图像视觉重叠有限的稀疏视图设定下，直接从查询与K个位姿数据库视图的稀疏匹配预测查询关键点的3D坐标，为视觉定位提供了一种不依赖离线SfM重建的可复用方案。
原文依据 ON3R predicts 3D coordinates for matched query keypoints, supervised by database reprojection residuals and a monocular depth prior.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 ON3R，一种在线训练的神经回归器，用于解决稀疏视图无结构定位问题，其中数据库图像视觉重叠有限且没有预构建的3D地图。给定查询图像与K元组已定位数据库视图之间的任意稀疏匹配，ON3R 预测匹配查询关键点的3D坐标，并以数据库重投影残差和单目深度先验作为监督。随后，通过 P3P-RANSAC 估计查询的绝对位姿，并用轻量级光束法平差进行优化。在 MegaDepth、Cambridge Landmarks 以及 Aachen Day-Night 的稀疏化版本上，ON3R 优于现有方法。当数据极度稀疏时——我们关注 K≤10 幅数据库图像——ON3R 尤为有效。代码见 https://github.com/ludvigdillen/ON3R。
英文原文摘要 800 字符
We present ON3R, an online-trained neural regressor addressing sparse-view structureless localization, where database images have limited visual overlap and no prebuilt 3D map. Given any sparse matches between a query and a K-tuple of posed database views, ON3R predicts 3D coordinates for matched query keypoints, supervised by database reprojection residuals and a monocular depth prior. Afterwards, the absolute pose of the query is estimated via P3P-RANSAC and refined with lightweight bundle adjustment. Across MegaDepth, Cambridge Landmarks, and a sparsified version of Aachen Day-Night, ON3R outperforms existing methods. ON3R is particularly effective when the data is extremely sparse -- we focus on K\!\leq\!10 database images. The code is available at https://github.com/ludvigdillen/ON3R.