mlsanityData to 7 Oct 2026

EFEM: Equivariant Neural Field Expectation Maximization for 3D Object Segmentation Without Scene Supervision

arXiv 2303.15440 ↗First posted 27 Mar 2023cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

We introduce Equivariant Neural Field Expectation Maximization (EFEM), a simple, effective, and robust geometric algorithm that can segment objects in 3D scenes without annotations or training on scenes. We achieve such unsupervised segmentation by exploiting single object shape priors. We make two novel steps in that direction. First, we introduce equivariant shape representations to this problem to eliminate the complexity induced by the variation in object configuration. Second, we propose a novel EM algorithm that can iteratively refine segmentation masks using the equivariant shape prior. We collect a novel real dataset Chairs and Mugs that contains various object configurations and novel scenes in order to verify the effectiveness and robustness of our method. Experimental results demonstrate that our method achieves consistent and robust performance across different scenes where the (weakly) supervised methods may fail. Code and data available at https://www.cis.upenn.edu/~leijh/projects/efem

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 70 of its 91 references that are in the map. First posted on arXiv 27 Mar 2023.

Builds on

70 of 91 references.

References · 70

  1. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  2. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  3. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  5. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  6. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  7. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  8. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  9. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,354
  10. 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
  11. 2020Convolutional Occupancy Networkscited by 760
  12. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
38 more
  1. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 602
  2. 2020SAPIEN: A SimulAted Part-based Interactive ENvironmentcited by 584
  3. 2021Neural Fields in Visual Computing and Beyondcited by 539
  4. 2020PointContrast: Unsupervised Pre-training for 3D Point Cloud Understandingcited by 502
  5. 20203D-FRONT: 3D Furnished Rooms with layOuts and semaNTicscited by 375
  6. 2020Local Implicit Grid Representations for 3D Scenescited by 372
  7. 2020Implicit Functions in Feature Space for 3D Shape Reconstruction and Completioncited by 343
  8. 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 332
  9. 2020PointGroup: Dual-Set Point Grouping for 3D Instance Segmentationcited by 318
  10. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 304
  11. 2018On the Generalization of Equivariance and Convolution in Neural Networks to the Action of Compact Groupscited by 296
  12. 2018A General Theory of Equivariant CNNs on Homogeneous Spacescited by 223
  13. 20183D-SIS: 3D Semantic Instance Segmentation of RGB-D Scanscited by 216
  14. 2020Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contextscited by 210
  15. 2021Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulationcited by 204
  16. 2022SoftGroup for 3D Instance Segmentation on Point Cloudscited by 197
  17. 2020OccuSeg: Occupancy-aware 3D Instance Segmentationcited by 177
  18. 2017SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentationcited by 168
  19. 2019Learning Object Bounding Boxes for 3D Instance Segmentation on Point Cloudscited by 156
  20. 2018GSPN: Generative Shape Proposal Network for 3D Instance Segmentation in Point Cloudcited by 145
  21. 2021Hierarchical Aggregation for 3D Instance Segmentationcited by 126
  22. 2018Building Generalizable Agents with a Realistic and Rich 3D Environmentcited by 109
  23. 2021Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networkscited by 98
  24. 2019BAE-NET: Branched Autoencoder for Shape Co-Segmentationcited by 89
  25. 2021Equivariant Point Network for 3D Point Cloud Analysiscited by 81
  26. 20193D Instance Segmentation via Multi-Task Metric Learningcited by 79
  27. 2021Coordinate Independent Convolutional Networks -- Isometry and Gauge Equivariant Convolutions on Riemannian Manifoldscited by 72
  28. 2020SDF-SRN: Learning Signed Distance 3D Object Reconstruction from Static Imagescited by 68
  29. 2022Towards Implicit Text-Guided 3D Shape Generationcited by 68
  30. 2022Approximate Convex Decomposition for 3D Meshes with Collision-Aware Concavity and Tree Searchcited by 61
  31. 2022CaDeX: Learning Canonical Deformation Coordinate Space for Dynamic Surface Representation via Neural Homeomorphismcited by 56
  32. 2022Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learningcited by 54
  33. 2019Point Cloud Instance Segmentation using Probabilistic Embeddingscited by 52
  34. 20213D Shape Generation with Grid-based Implicit Functionscited by 48
  35. 2021MultiBodySync: Multi-Body Segmentation and Motion Estimation via 3D Scan Synchronizationcited by 46
  36. 2021Leveraging SE(3) Equivariance for Self-Supervised Category-Level Object Pose Estimationcited by 45
  37. 2020RfD-Net: Point Scene Understanding by Semantic Instance Reconstructioncited by 36
  38. 20223D Instances as 1D Kernelscited by 36

50 most cited of 70 All 70 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 7; 2024: 5; 2025: 4; 2026: 2.

18 papers.

Sort:

By area:

  1. 2024EquiBot: SIM(3)-Equivariant Diffusion Policy for Generalizable and Data Efficient Learningcited by 72
  2. 2023Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulationcited by 51
  3. 2023EquivAct: SIM(3)-Equivariant Visuomotor Policies beyond Rigid Object Manipulationcited by 37
  4. 2023Banana: Banach Fixed-Point Network for Pointcloud Segmentation with Inter-Part Equivariancecited by 19
  5. 2023ZeroPS: High-quality Cross-modal Knowledge Transfer for Zero-Shot 3D Part Segmentationcited by 18
  6. 2023U3DS3: Unsupervised 3D Semantic Scene Segmentationcited by 16
  7. 2025A Survey of Robotic Navigation and Manipulation with Physics Simulators in the Era of Embodied AIcited by 16
  8. 2023Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environmentscited by 15
  9. 2024RiEMann: Near Real-Time SE(3)-Equivariant Robot Manipulation without Point Cloud Segmentationcited by 12
  10. 2023Multi-body SE(3) Equivariance for Unsupervised Rigid Segmentation and Motion Estimationcited by 7
  11. 2025GAGrasp: Geometric Algebra Diffusion for Dexterous Graspingcited by 7
  12. 2024EqvAfford: SE(3) Equivariance for Point-Level Affordance Learningcited by 6
6 more
  1. 2025GrabS: Generative Embodied Agent for 3D Object Segmentation without Scene Supervisioncited by 3
  2. 2024Current Symmetry Group Equivariant Convolution Frameworks for Representation Learningcited by 2
  3. 2025Learning Generalizable Shape Completion with SIM(3) Equivariancecited by 2
  4. 2024Bayesian Self-Training for Semi-Supervised 3D Segmentationcited by 1
  5. 2026EvObj: Learning Evolving Object-centric Representations for 3D Instance Segmentation without Scene Supervisioncited by 1
  6. 2026FoundObj: Self-supervised Foundation Models as Rewards for Label-free 3D Object Segmentationcited by 0