We study the problem of extracting accurate correspondences for point cloud registration. Recent keypoint-free methods have shown great potential through bypassing the detection of repeatable keypoints which is difficult to do especially in low-overlap scenarios. They seek correspondences over downsampled superpoints, which are then propagated to dense points. Superpoints are matched based on whether their neighboring patches overlap. Such sparse and loose matching requires contextual features capturing the geometric structure of the point clouds. We propose Geometric Transformer, or GeoTransformer for short, to learn geometric feature for robust superpoint matching. It encodes pair-wise distances and triplet-wise angles, making it invariant to rigid transformation and robust in low-overlap cases. The simplistic design attains surprisingly high matching accuracy such that no RANSAC is required in the estimation of alignment transformation, leading to 100 times acceleration. Extensive experiments on rich benchmarks encompassing indoor, outdoor, synthetic, multiway and non-rigid demonstrate the efficacy of GeoTransformer. Notably, our method improves the inlier ratio by <inline-formula><tex-math notation="LaTeX">$18{\sim }31$</tex-math><alternatives><mml:math><mml:mrow><mml:mn>18</mml:mn><mml:mo>∼</mml:mo><mml:mn>31</mml:mn></mml:mrow></mml:math><inline-graphic xlink:href="qin-ieq1-3259038.gif"/></alternatives></inline-formula> percentage points and the registration recall by over 7 points on the challenging 3DLoMatch benchmark.
GeoTransformer: Fast and Robust Point Cloud Registration With Geometric Transformer
Zheng Qin,Hao Yu,Changjian Wang,Yulan Guo,Yuxing Peng,Slobodan Ilic,Dewen Hu,Kai Xu
Published 2023 in IEEE Transactions on Pattern Analysis and Machine Intelligence
ABSTRACT
PUBLICATION RECORD
- Publication year
2023
- Venue
IEEE Transactions on Pattern Analysis and Machine Intelligence
- Publication date
2023-03-20
- Fields of study
Medicine, Computer Science, Engineering
- Identifiers
- External record
- Source metadata
Semantic Scholar, PubMed
CITATION MAP
EXTRACTION MAP
CLAIMS
- No claims are published for this paper.
CONCEPTS
- No concepts are published for this paper.
REFERENCES
Showing 1-64 of 64 references · Page 1 of 1