Equivariant Flow Matching for Point Cloud Assembly

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Ziming, Xue, Nan, Jörnsten, Rebecka
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915751448805376
author Wang, Ziming
Xue, Nan
Jörnsten, Rebecka
author_facet Wang, Ziming
Xue, Nan
Jörnsten, Rebecka
contents The goal of point cloud assembly is to reconstruct a complete 3D shape by aligning multiple point cloud pieces. This work presents a novel equivariant solver for assembly tasks based on flow matching models. We first theoretically show that the key to learning equivariant distributions via flow matching is to learn related vector fields. Based on this result, we propose an assembly model, called equivariant diffusion assembly (Eda), which learns related vector fields conditioned on the input pieces. We further construct an equivariant path for Eda, which guarantees high data efficiency of the training process. Our numerical results show that Eda is highly competitive on practical datasets, and it can even handle the challenging situation where the input pieces are non-overlapped.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21539
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Equivariant Flow Matching for Point Cloud Assembly
Wang, Ziming
Xue, Nan
Jörnsten, Rebecka
Computer Vision and Pattern Recognition
Artificial Intelligence
The goal of point cloud assembly is to reconstruct a complete 3D shape by aligning multiple point cloud pieces. This work presents a novel equivariant solver for assembly tasks based on flow matching models. We first theoretically show that the key to learning equivariant distributions via flow matching is to learn related vector fields. Based on this result, we propose an assembly model, called equivariant diffusion assembly (Eda), which learns related vector fields conditioned on the input pieces. We further construct an equivariant path for Eda, which guarantees high data efficiency of the training process. Our numerical results show that Eda is highly competitive on practical datasets, and it can even handle the challenging situation where the input pieces are non-overlapped.
title Equivariant Flow Matching for Point Cloud Assembly
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2505.21539