SE3ET: SE(3)-Equivariant Transformer for Low-Overlap Point Cloud Registration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Chien Erh, Zhu, Minghan, Ghaffari, Maani
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910539209244672
author Lin, Chien Erh
Zhu, Minghan
Ghaffari, Maani
author_facet Lin, Chien Erh
Zhu, Minghan
Ghaffari, Maani
contents Partial point cloud registration is a challenging problem in robotics, especially when the robot undergoes a large transformation, causing a significant initial pose error and a low overlap between measurements. This work proposes exploiting equivariant learning from 3D point clouds to improve registration robustness. We propose SE3ET, an SE(3)-equivariant registration framework that employs equivariant point convolution and equivariant transformer designs to learn expressive and robust geometric features. We tested the proposed registration method on indoor and outdoor benchmarks where the point clouds are under arbitrary transformations and low overlapping ratios. We also provide generalization tests and run-time performance.
format Preprint
id arxiv_https___arxiv_org_abs_2407_16823
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SE3ET: SE(3)-Equivariant Transformer for Low-Overlap Point Cloud Registration
Lin, Chien Erh
Zhu, Minghan
Ghaffari, Maani
Robotics
Partial point cloud registration is a challenging problem in robotics, especially when the robot undergoes a large transformation, causing a significant initial pose error and a low overlap between measurements. This work proposes exploiting equivariant learning from 3D point clouds to improve registration robustness. We propose SE3ET, an SE(3)-equivariant registration framework that employs equivariant point convolution and equivariant transformer designs to learn expressive and robust geometric features. We tested the proposed registration method on indoor and outdoor benchmarks where the point clouds are under arbitrary transformations and low overlapping ratios. We also provide generalization tests and run-time performance.
title SE3ET: SE(3)-Equivariant Transformer for Low-Overlap Point Cloud Registration
topic Robotics
url https://arxiv.org/abs/2407.16823