FRIDU: Functional Map Refinement with Guided Image Diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rimon, Avigail Cohen, Ben-Chen, Mirela, Litany, Or
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918061517307904
author Rimon, Avigail Cohen
Ben-Chen, Mirela
Litany, Or
author_facet Rimon, Avigail Cohen
Ben-Chen, Mirela
Litany, Or
contents We propose a novel approach for refining a given correspondence map between two shapes. A correspondence map represented as a functional map, namely a change of basis matrix, can be additionally treated as a 2D image. With this perspective, we train an image diffusion model directly in the space of functional maps, enabling it to generate accurate maps conditioned on an inaccurate initial map. The training is done purely in the functional space, and thus is highly efficient. At inference time, we use the pointwise map corresponding to the current functional map as guidance during the diffusion process. The guidance can additionally encourage different functional map objectives, such as orthogonality and commutativity with the Laplace-Beltrami operator. We show that our approach is competitive with state-of-the-art methods of map refinement and that guided diffusion models provide a promising pathway to functional map processing.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14322
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FRIDU: Functional Map Refinement with Guided Image Diffusion
Rimon, Avigail Cohen
Ben-Chen, Mirela
Litany, Or
Computer Vision and Pattern Recognition
Machine Learning
We propose a novel approach for refining a given correspondence map between two shapes. A correspondence map represented as a functional map, namely a change of basis matrix, can be additionally treated as a 2D image. With this perspective, we train an image diffusion model directly in the space of functional maps, enabling it to generate accurate maps conditioned on an inaccurate initial map. The training is done purely in the functional space, and thus is highly efficient. At inference time, we use the pointwise map corresponding to the current functional map as guidance during the diffusion process. The guidance can additionally encourage different functional map objectives, such as orthogonality and commutativity with the Laplace-Beltrami operator. We show that our approach is competitive with state-of-the-art methods of map refinement and that guided diffusion models provide a promising pathway to functional map processing.
title FRIDU: Functional Map Refinement with Guided Image Diffusion
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2506.14322