SGMatch: Semantic-Guided Non-Rigid Shape Matching with Flow Regularization

TL;DR

SGMatch is a learning-based framework that improves non-rigid 3D shape matching by integrating semantic features and flow regularization, achieving better accuracy under challenging deformations and noise.

Contribution

It introduces a semantic-guided local cross-attention module and a flow regularization objective, advancing non-rigid shape matching methods.

Findings

Achieves competitive performance on near-isometric shapes.

Shows improved accuracy under non-isometric deformations.

Handles topological noise effectively.

Abstract

Establishing accurate point-to-point correspondences between non-rigid 3D shapes remains a critical challenge, particularly under non-isometric deformations and topological noise. Existing functional map pipelines suffer from ambiguities that geometric descriptors alone cannot resolve, and spatial inconsistencies inherent in the projection of truncated spectral bases to dense pointwise correspondences. In this paper, we introduce SGMatch, a learning-based framework for semantic-guided non-rigid shape matching. Specifically, we design a Semantic-Guided Local Cross-Attention module that integrates semantic features from vision foundation models into geometric descriptors while preserving local structural continuity. Furthermore, we introduce a regularization objective based on conditional flow matching, which supervises a time-varying velocity field to encourage spatial smoothness of the recovered correspondences. Experimental results on multiple benchmarks demonstrate that SGMatch achieves competitive performance across near-isometric settings and consistent improvements under non-isometric deformations and topological noise.