CoFiNet: Reliable Coarse-to-fine Correspondences for Robust Point Cloud Registration
This addresses robust point cloud registration for applications like robotics and 3D reconstruction, representing a novel method for a known bottleneck.
The paper tackles the problem of unreliable keypoint repeatability in point cloud registration by introducing CoFiNet, a coarse-to-fine network that extracts hierarchical correspondences without keypoint detection, achieving at least 5% higher Registration Recall on the 3DLoMatch benchmark with fewer parameters.
We study the problem of extracting correspondences between a pair of point clouds for registration. For correspondence retrieval, existing works benefit from matching sparse keypoints detected from dense points but usually struggle to guarantee their repeatability. To address this issue, we present CoFiNet - Coarse-to-Fine Network which extracts hierarchical correspondences from coarse to fine without keypoint detection. On a coarse scale and guided by a weighting scheme, our model firstly learns to match down-sampled nodes whose vicinity points share more overlap, which significantly shrinks the search space of a consecutive stage. On a finer scale, node proposals are consecutively expanded to patches that consist of groups of points together with associated descriptors. Point correspondences are then refined from the overlap areas of corresponding patches, by a density-adaptive matching module capable to deal with varying point density. Extensive evaluation of CoFiNet on both indoor and outdoor standard benchmarks shows our superiority over existing methods. Especially on 3DLoMatch where point clouds share less overlap, CoFiNet significantly outperforms state-of-the-art approaches by at least 5% on Registration Recall, with at most two-third of their parameters.