Deep-PE: A Learning-Based Pose Evaluator for Point Cloud Registration

TL;DR

Deep-PE is a novel learning-based pose evaluation method for point cloud registration that improves accuracy in low-overlap scenarios by predicting pose success likelihood without relying on explicit correspondences.

Contribution

It introduces Deep-PE with Pose-Aware Attention and Pose Confidence Prediction modules, pioneering deep learning for pose selection without explicit correspondences in point cloud registration.

Findings

Outperforms state-of-the-art methods by 8-11% in registration recall on 3DLoMatch.

Effective in low-overlap scenarios, enhancing registration accuracy.

First to use deep learning for pose selection without explicit correspondences.

Abstract

In the realm of point cloud registration, the most prevalent pose evaluation approaches are statistics-based, identifying the optimal transformation by maximizing the number of consistent correspondences. However, registration recall decreases significantly when point clouds exhibit a low overlap rate, despite efforts in designing feature descriptors and establishing correspondences. In this paper, we introduce Deep-PE, a lightweight, learning-based pose evaluator designed to enhance the accuracy of pose selection, especially in challenging point cloud scenarios with low overlap. Our network incorporates a Pose-Aware Attention (PAA) module to simulate and learn the alignment status of point clouds under various candidate poses, alongside a Pose Confidence Prediction (PCP) module that predicts the likelihood of successful registration. These two modules facilitate the learning of both local and global alignment priors. Extensive tests across multiple benchmarks confirm the effectiveness of Deep-PE. Notably, on 3DLoMatch with a low overlap rate, Deep-PE significantly outperforms state-of-the-art methods by at least 8% and 11% in registration recall under handcrafted FPFH and learning-based FCGF descriptors, respectively. To the best of our knowledge, this is the first study to utilize deep learning to select the optimal pose without the explicit need for input correspondences.