Register Any Point: Scaling 3D Point Cloud Registration by Flow Matching
Abstract
Point cloud registration is formulated as conditional generation using a velocity field to transport noisy points into a common frame, enabling accurate multi-view registration with low overlap and cross-scale generalization.
Point cloud registration aligns multiple unposed point clouds into a common frame, and is a core step for 3D reconstruction and robot localization. In this work, we cast registration as conditional generation: a learned continuous, point-wise velocity field transports noisy points to a registered scene, from which the pose of each view is recovered. Unlike previous methods that conduct correspondence matching to estimate the transformation between a pair of point clouds and then optimize the pairwise transformations to realize multi-view registration, our model directly generates the registered point cloud. With a lightweight local feature extractor and test-time rigidity enforcement, our approach achieves state-of-the-art results on pairwise and multi-view registration benchmarks, particularly with low overlap, and generalizes across scales and sensor modalities. It further supports downstream tasks including relocalization, multi-robot SLAM, and multi-session map merging. Source code available at: https://github.com/PRBonn/RAP.
Models citing this paper 1
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper