AI RESEARCH

Register Any Point: Scaling 3D Point Cloud Registration by Flow Matching

arXiv CS.CV

ArXi:2512.01850v2 Announce Type: replace Point cloud registration aligns multiple unposed point clouds into a common reference frame and is a core step for 3D reconstruction and robot localization without initial guess. 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.