A Robust Loss for Point Cloud Registration

TL;DR

This paper introduces a new robust metric for point cloud registration based on shape intersections with random lines, improving accuracy and stability over traditional closest-point methods.

Contribution

The paper proposes a novel intersection-based metric for surface registration that does not rely on point correspondences, enhancing robustness and performance.

Findings

Outperforms state-of-the-art methods in registration accuracy

Effective in both direct optimization and unsupervised learning

Demonstrates robustness against correspondence instability

Abstract

The performance of surface registration relies heavily on the metric used for the alignment error between the source and target shapes. Traditionally, such a metric is based on the point-to-point or point-to-plane distance from the points on the source surface to their closest points on the target surface, which is susceptible to failure due to instability of the closest-point correspondence. In this paper, we propose a novel metric based on the intersection points between the two shapes and a random straight line, which does not assume a specific correspondence. We verify the effectiveness of this metric by extensive experiments, including its direct optimization for a single registration problem as well as unsupervised learning for a set of registration problems. The results demonstrate that the algorithms utilizing our proposed metric outperforms the state-of-the-art optimization-based and unsupervised learning-based methods.