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

TL;DR

Q-REG introduces an end-to-end trainable point cloud registration method that leverages surface curvature for robust rigid pose estimation, improving accuracy across multiple benchmarks.

Contribution

It proposes a novel end-to-end training framework for point cloud registration that integrates geometric surface information into the pose estimation process.

Findings

Sets new state-of-the-art on 3DMatch, KITTI, and ModelNet benchmarks.

Demonstrates robustness and consistency when integrated with various correspondence matching methods.

Enables fully end-to-end training by formalizing pose estimation as an exhaustive search.

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.