GETr: A Geometric Equivariant Transformer for Point Cloud Registration
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association and John Wiley & Sons Ltd.
Abstract
As a fundamental problem in computer vision, 3D point cloud registration (PCR) aims to seek the optimal transformation to align point cloud pairs. Meanwhile, the equivariance lies at the core of matching point clouds at arbitrary pose. In this paper, we propose GETr, a geometric equivariant transformer for PCR. By learning the point-wise orientations, we decouple the coordinate to the pose of the point clouds, which is the key to achieve equivariance in our framework. Then we utilize attention mechanism to learn the geometric features for superpoints matching, the proposed novel self-attention mechanism encodes the geometric information of point clouds. Finally, the coarse-to-fine manner is used to obtain high-quality correspondence for registration. Extensive experiments on both indoor and outdoor benchmarks demonstrate that our method outperforms various existing state-of-the-art methods.
Description
CCS Concepts: Computing methodologies->Computer graphics; Computer vision
@article{10.1111:cgf.15216,
journal = {Computer Graphics Forum},
title = {{GETr: A Geometric Equivariant Transformer for Point Cloud Registration}},
author = {Yu, Chang and Zhang, Sanguo and Shen, Li-Yong},
year = {2024},
publisher = {The Eurographics Association and John Wiley & Sons Ltd.},
ISSN = {1467-8659},
DOI = {10.1111/cgf.15216}
}