English

Q-REG: End-to-End Trainable Point Cloud Registration with Surface Curvature

Computer Vision and Pattern Recognition 2023-09-29 v1

Abstract

Point cloud registration has seen recent success with several learning-based methods that focus on correspondence matching and, as such, optimize only for this objective. Following the learning step of correspondence matching, they evaluate the estimated rigid transformation with a RANSAC-like framework. While it is an indispensable component of these methods, it prevents a fully end-to-end training, leaving the objective to minimize the pose error nonserved. We present a novel solution, Q-REG, which utilizes rich geometric information to estimate the rigid pose from a single correspondence. Q-REG allows to formalize the robust estimation as an exhaustive search, hence enabling end-to-end training that optimizes over both objectives of correspondence matching and rigid pose estimation. We demonstrate in the experiments that Q-REG is agnostic to the correspondence matching method and provides consistent improvement both when used only in inference and in end-to-end training. It sets a new state-of-the-art on the 3DMatch, KITTI, and ModelNet benchmarks.

Keywords

Cite

@article{arxiv.2309.16023,
  title  = {Q-REG: End-to-End Trainable Point Cloud Registration with Surface Curvature},
  author = {Shengze Jin and Daniel Barath and Marc Pollefeys and Iro Armeni},
  journal= {arXiv preprint arXiv:2309.16023},
  year   = {2023}
}
R2 v1 2026-06-28T12:34:20.785Z