mlsanityData to 7 Oct 2026

Adjoint Rigid Transform Network: Task-conditioned Alignment of 3D Shapes

arXiv 2102.01161 ↗First posted 1 Feb 2021cs.CVcited by 4

Not on the map. Why?

Abstract (arXiv)

Most learning methods for 3D data (point clouds, meshes) suffer significant performance drops when the data is not carefully aligned to a canonical orientation. Aligning real world 3D data collected from different sources is non-trivial and requires manual intervention. In this paper, we propose the Adjoint Rigid Transform (ART) Network, a neural module which can be integrated with a variety of 3D networks to significantly boost their performance. ART learns to rotate input shapes to a learned canonical orientation, which is crucial for a lot of tasks such as shape reconstruction, interpolation, non-rigid registration, and latent disentanglement. ART achieves this with self-supervision and a rotation equivariance constraint on predicted rotations. The remarkable result is that with only self-supervision, ART facilitates learning a unique canonical orientation for both rigid and nonrigid shapes, which leads to a notable boost in performance of aforementioned tasks. We will release our code and pre-trained models for further research.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 36 of its 58 references that are in the map. First posted on arXiv 1 Feb 2021.

Builds on

36 of 58 references.

References · 36

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  4. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  5. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  6. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  7. 2015Spatial Transformer Networkscited by 1,423
  8. 2019AMASS: Archive of Motion Capture as Surface Shapescited by 1,208
  9. 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,186
  10. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
  11. 2019General E(2)-Equivariant Steerable CNNscited by 441
  12. 20163D Menagerie: Modeling the 3D shape and pose of animalscited by 310
24 more
  1. 2016Steerable CNNscited by 298
  2. 2018Generating 3D faces using Convolutional Mesh Autoencoderscited by 267
  3. 2017Learning SO(3) Equivariant Representations with Spherical CNNscited by 244
  4. 2017Learning Steerable Filters for Rotation Equivariant CNNscited by 193
  5. 20183D-CODED : 3D Correspondences by Deep Deformationcited by 160
  6. 2020Combining Implicit Function Learning and Parametric Models for 3D Human Reconstructioncited by 133
  7. 2020LoopReg: Self-supervised Learning of Implicit Surface Correspondences, Pose and Shape for 3D Human Mesh Registrationcited by 108
  8. 2016Exploiting Cyclic Symmetry in Convolutional Neural Networkscited by 91
  9. 2019Neural 3D Morphable Models: Spiral Convolutional Networks for 3D Shape Representation Learning and Generationcited by 89
  10. 2019SpiralNet++: A Fast and Highly Efficient Mesh Convolution Operatorcited by 89
  11. 2017Oriented Response Networkscited by 67
  12. 2020Unsupervised Shape and Pose Disentanglement for 3D Meshescited by 61
  13. 2015Building Statistical Shape Spaces for 3D Human Modelingcited by 47
  14. 20193D-Rotation-Equivariant Quaternion Neural Networkscited by 30
  15. 2020Canonical Capsules: Self-Supervised Capsules in Canonical Posecited by 30
  16. 2016Inverse Compositional Spatial Transformer Networkscited by 29
  17. 2020Learning Implicit Functions for Topology-Varying Dense 3D Shape Correspondencecited by 27
  18. 2020Learning to Orient Surfaces by Self-supervised Spherical CNNscited by 26
  19. 2020Image Morphing with Perceptual Constraints and STN Alignmentcited by 13
  20. 20173D Morphable Models as Spatial Transformer Networkscited by 12
  21. 2018Iterative Transformer Network for 3D Point Cloudcited by 11
  22. 2020When Deep Learning Meets Data Alignment: A Review on Deep Registration Networks (DRNs)cited by 4
  23. 2020Spatial Transformer Point Convolutioncited by 3
  24. 2018DeSTNet: Densely Fused Spatial Transformer Networkscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 2; 2025: 0; 2026: 1.

4 papers.

  1. 2023Template Free Reconstruction of Human-object Interaction with Procedural Interaction Generationcited by 37
  2. 2024InterTrack: Tracking Human Object Interaction without Object Templatescited by 22
  3. 2026Unleashing Guidance Without Classifiers for Human-Object Interaction Animationcited by 5
  4. 2024Symmetry-Robust 3D Orientation Estimationcited by 1