DeepShapeMatchingKit: Accelerated Functional Map Solver and Shape Matching Pipelines Revisited

2026-04-11

AI Summary

This work addresses the inefficiency, implementation inconsistency, and limited evaluation metrics in functional map computation for non-rigid 3D shape matching. We propose a vectorized functional map solver that efficiently resolves all linear systems via a single kernel call, achieving up to a 33× speedup while preserving solution accuracy. Furthermore, we unify the training, evaluation, and data pipeline for deep shape matching, elucidate the distinct behaviors of two spatial gradient variants in DiffusionNet, and introduce “balanced accuracy” as a novel metric for partial-to-partial correspondence evaluation. Extensive experiments across multiple benchmarks validate the effectiveness and efficiency of our approach. The implementation is publicly released as the DeepShapeMatchingKit toolkit.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning with Manifolds
• Intelligent Robots: Localization, Mapping, and Navigation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Deep functional maps, leveraging learned feature extractors and spectral correspondence solvers, are fundamental to non-rigid 3D shape matching. Based on an analysis of open-source implementations, we find that standard functional map implementations solve k independent linear systems serially, which is a computational bottleneck at higher spectral resolution. We thus propose a vectorized reformulation that solves all systems in a single kernel call, achieving up to a 33x speedup while preserving the exact solution. Furthermore, we identify and document a previously unnoticed implementation divergence in the spatial gradient features of the mainstay DiffusionNet: two variants that parameterize distinct families of tangent-plane transformations, and present experiments analyzing their respective behaviors across diverse benchmarks. We additionally revisit overlap prediction evaluation for partial-to-partial matching and show that balanced accuracy provides a useful complementary metric under varying overlap ratios. To share these advancements with the wider community, we present an open-source codebase, DeepShapeMatchingKit, that incorporates these improvements and standardizes training, evaluation, and data pipelines for common deep shape matching methods. The codebase is available at: https://github.com/xieyizheng/DeepShapeMatchingKit

Problem

• functional maps
• shape matching
• computational bottleneck
• implementation divergence
• overlap prediction

Innovation

• functional maps
• vectorized solver
• DiffusionNet
• shape matching
• balanced accuracy