mlsanityData to 7 Oct 2026

Coarse-to-Fine Point Cloud Registration with SE(3)-Equivariant Representations

arXiv 2210.02045 ↗First posted 5 Oct 2022cs.CV cs.ROcited by 9

Not on the map. Why?

Abstract (arXiv)

Point cloud registration is a crucial problem in computer vision and robotics. Existing methods either rely on matching local geometric features, which are sensitive to the pose differences, or leverage global shapes, which leads to inconsistency when facing distribution variances such as partial overlapping. Combining the advantages of both types of methods, we adopt a coarse-to-fine pipeline that concurrently handles both issues. We first reduce the pose differences between input point clouds by aligning global features; then we match the local features to further refine the inaccurate alignments resulting from distribution variances. As global feature alignment requires the features to preserve the poses of input point clouds and local feature matching expects the features to be invariant to these poses, we propose an SE(3)-equivariant feature extractor to simultaneously generate two types of features. In this feature extractor, representations that preserve the poses are first encoded by our novel SE(3)-equivariant network and then converted into pose-invariant ones by a pose-detaching module. Experiments demonstrate that our proposed method increases the recall rate by 20% compared to state-of-the-art methods when facing both pose differences and distribution variances.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 33 of its 48 references that are in the map. First posted on arXiv 5 Oct 2022.

Builds on

33 of 48 references.

References · 33

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  2. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  3. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 602
  4. 2019Deep Closest Point: Learning Representations for Point Cloud Registrationcited by 318
  5. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 304
  6. 2016Steerable CNNscited by 298
  7. 2018On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groupscited by 296
  8. 2020PREDATOR: Registration of 3D Point Clouds with Low Overlapcited by 287
  9. 2020Generalizing Convolutional Neural Networks for Equivariance to Lie Groups on Arbitrary Continuous Datacited by 248
  10. 2017Learning SO(3) Equivariant Representations with Spherical CNNscited by 244
  11. 2020Deep Global Registrationcited by 228
  12. 2021Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulationcited by 204
21 more
  1. 2022Geometric Transformer for Fast and Robust Point Cloud Registrationcited by 200
  2. 2020RPM-Net: Robust Point Matching using Learned Featurescited by 181
  3. 2018PPFNet: Global Context Aware Local Features for Robust 3D Point Matchingcited by 180
  4. 2020D3Feat: Joint Learning of Dense Detection and Description of 3D Local Featurescited by 175
  5. 2018The Perfect Match: 3D Point Cloud Matching with Smoothed Densitiescited by 169
  6. 2016Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registrationcited by 159
  7. 2019PRNet: Self-Supervised Learning for Partial-to-Partial Registrationcited by 146
  8. 2020SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registrationcited by 141
  9. 2022REGTR: End-to-end Point Cloud Correspondences with Transformerscited by 94
  10. 2020DeepGMR: Learning Latent Gaussian Mixture Models for Registrationcited by 91
  11. 2019DeepICP: An End-to-End Deep Neural Network for 3D Point Cloud Registrationcited by 77
  12. 2021LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAMcited by 67
  13. 2018Learning 3D Shape Completion under Weak Supervisioncited by 40
  14. 2021Correspondence-Free Point Cloud Registration with SO(3)-Equivariant Implicit Shape Representationscited by 24
  15. 2021ReAgent: Point Cloud Registration using Imitation and Reinforcement Learningcited by 23
  16. 2016GOGMA: Globally-Optimal Gaussian Mixture Alignmentcited by 22
  17. 2018The Alignment of the Spheres: Globally-Optimal Spherical Mixture Alignment for Camera Pose Estimationcited by 20
  18. 2022ImLoveNet: Misaligned Image-supported Registration Network for Low-overlap Point Cloud Pairscited by 17
  19. 2021DetarNet: Decoupling Translation and Rotation by Siamese Network for Point Cloud Registrationcited by 12
  20. 2014Convex Relaxations of SE(2) and SE(3) for Visual Pose Estimationcited by 3
  21. 2020Best Buddies Registration for Point Cloudscited by 3

Built on by

Cited by year

Citing papers per year: 2023: 4; 2024: 4; 2025: 1.

9 papers.

  1. 2023Leveraging SE(3) Equivariance for Learning 3D Geometric Shape Assemblycited by 27
  2. 2023Banana: Banach Fixed-Point Network for Pointcloud Segmentation with Inter-Part Equivariancecited by 19
  3. 2023EFEM: Equivariant Neural Field Expectation Maximization for 3D Object Segmentation Without Scene Supervisioncited by 18
  4. 2023NeuSE: Neural SE(3)-Equivariant Embedding for Consistent Spatial Understanding with Objectscited by 10
  5. 2024Approximately Piecewise E(3) Equivariant Point Networkscited by 4
  6. 2024SE(3)-bi-equivariant Transformers for Point Cloud Assemblycited by 4
  7. 2024Rigid Single-Slice-in-Volume registration via rotation-equivariant 2D/3D feature matchingcited by 4
  8. 2024BYE: Build Your Encoder with One Sequence of Exploration Data for Long-Term Dynamic Scene Understandingcited by 2
  9. 2025Learning Generalizable Shape Completion with SIM(3) Equivariancecited by 2