SRFeat: Learning Locally Accurate and Globally Consistent Non-Rigid Shape Correspondence

TL;DR

This paper introduces a learning framework that combines contrastive feature learning with geometric regularization to achieve accurate and consistent non-rigid shape and image keypoint matching.

Contribution

It proposes a novel combination of contrastive learning and geometric regularization to improve correspondence accuracy and consistency in non-rigid matching tasks.

Findings

Outperforms existing methods on 3D non-rigid shape correspondence benchmarks.

Achieves state-of-the-art results in 2D image keypoint matching.

Produces highly discriminative and smooth features for robust matching.

Abstract

In this work, we present a novel learning-based framework that combines the local accuracy of contrastive learning with the global consistency of geometric approaches, for robust non-rigid matching. We first observe that while contrastive learning can lead to powerful point-wise features, the learned correspondences commonly lack smoothness and consistency, owing to the purely combinatorial nature of the standard contrastive losses. To overcome this limitation we propose to boost contrastive feature learning with two types of smoothness regularization that inject geometric information into correspondence learning. With this novel combination in hand, the resulting features are both highly discriminative across individual points, and, at the same time, lead to robust and consistent correspondences, through simple proximity queries. Our framework is general and is applicable to local feature learning in both the 3D and 2D domains. We demonstrate the superiority of our approach through extensive experiments on a wide range of challenging matching benchmarks, including 3D non-rigid shape correspondence and 2D image keypoint matching.