mlsanityData to 7 Oct 2026

MATCH POLICY: A Simple Pipeline from Point Cloud Registration to Manipulation Policies

arXiv 2409.15517 ↗First posted 23 Sep 2024cs.RO cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Many manipulation tasks require the robot to rearrange objects relative to one another. Such tasks can be described as a sequence of relative poses between parts of a set of rigid bodies. In this work, we propose MATCH POLICY, a simple but novel pipeline for solving high-precision pick and place tasks. Instead of predicting actions directly, our method registers the pick and place targets to the stored demonstrations. This transfers action inference into a point cloud registration task and enables us to realize nontrivial manipulation policies without any training. MATCH POLICY is designed to solve high-precision tasks with a key-frame setting. By leveraging the geometric interaction and the symmetries of the task, it achieves extremely high sample efficiency and generalizability to unseen configurations. We demonstrate its state-of-the-art performance across various tasks on RLBench benchmark compared with several strong baselines and test it on a real robot with six tasks.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 59 of its 80 references that are in the map. First posted on arXiv 23 Sep 2024.

Builds on

59 of 80 references.

References · 59

  1. 2023Segment Anythingcited by 8,158
  2. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  3. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  4. 2018Open3D: A Modern Library for 3D Data Processingcited by 717
  5. 2019RLBench: The Robot Learning Benchmark & Learning Environmentcited by 669
  6. 2022Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulationcited by 545
  7. 2019General E(2)-Equivariant Steerable CNNscited by 441
  8. 2023Segment Anything in High Qualitycited by 373
  9. 2019Deep Closest Point: Learning Representations for Point Cloud Registrationcited by 318
  10. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 304
  11. 2020PREDATOR: Registration of 3D Point Clouds with Low Overlapcited by 287
  12. 20243D Diffuser Actor: Policy Diffusion with 3D Scene Representationscited by 278
38 more
  1. 2022Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphscited by 258
  2. 2023RVT: Robotic View Transformer for 3D Object Manipulationcited by 224
  3. 2021Neural Descriptor Fields: SE(3)-Equivariant Object Representations for Manipulationcited by 204
  4. 2016Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registrationcited by 159
  5. 2019PRNet: Self-Supervised Learning for Partial-to-Partial Registrationcited by 146
  6. 2023Act3D: 3D Feature Field Transformers for Multi-Task Robotic Manipulationcited by 135
  7. 20243D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representationscited by 115
  8. 2024Hierarchical Diffusion Policy for Kinematics-Aware Multi-Task Robotic Manipulationcited by 92
  9. 2021A comprehensive survey on point cloud registrationcited by 84
  10. 2022SO(2)-Equivariant Reinforcement Learningcited by 79
  11. 2022DexPoint: Generalizable Point Cloud Reinforcement Learning for Sim-to-Real Dexterous Manipulationcited by 78
  12. 2024EquiBot: SIM(3)-Equivariant Diffusion Policy for Generalizable and Data Efficient Learningcited by 72
  13. 2020Fast and Robust Iterative Closest Pointcited by 67
  14. 2021Equivariant Q Learning in Spatial Action Spacescited by 61
  15. 2024Equivariant Diffusion Policycited by 61
  16. 2022Fast Point Cloud Generation with Straight Flowscited by 56
  17. 2023PolarNet: 3D Point Clouds for Language-Guided Robotic Manipulationcited by 56
  18. 2022Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learningcited by 54
  19. 2022Sample Efficient Grasp Learning Using Equivariant Modelscited by 52
  20. 2022SE(3)-Equivariant Relational Rearrangement with Neural Descriptor Fieldscited by 51
  21. 2023Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulationcited by 51
  22. 2022TAX-Pose: Task-Specific Cross-Pose Estimation for Robot Manipulationcited by 47
  23. 2023GenPose: Generative Category-level Object Pose Estimation via Diffusion Modelscited by 47
  24. 2024Point Cloud Matters: Rethinking the Impact of Different Observation Spaces on Robot Learningcited by 45
  25. 2022Equivariant Transporter Networkcited by 44
  26. 2022Neural Shape Mating: Self-Supervised Object Assembly with Adversarial Shape Priorscited by 42
  27. 2024RISE: 3D Perception Makes Real-World Robot Imitation Simple and Effectivecited by 41
  28. 2022On-Robot Learning With Equivariant Modelscited by 38
  29. 2023EquivAct: SIM(3)-Equivariant Visuomotor Policies beyond Rigid Object Manipulationcited by 37
  30. 2022Edge Grasp Network: A Graph-Based SE(3)-invariant Approach to Grasp Detectioncited by 34
  31. 2024Fourier Transporter: Bi-Equivariant Robotic Manipulation in 3Dcited by 32
  32. 2022Frame Mining: a Free Lunch for Learning Robotic Manipulation from 3D Point Cloudscited by 31
  33. 2023Shelving, Stacking, Hanging: Relational Pose Diffusion for Multi-modal Rearrangementcited by 29
  34. 2019SDRSAC: Semidefinite-Based Randomized Approach for Robust Point Cloud Registration without Correspondencescited by 26
  35. 2022SEIL: Simulation-augmented Equivariant Imitation Learningcited by 22
  36. 2024Deep SE(3)-Equivariant Geometric Reasoning for Precise Placement Taskscited by 19
  37. 2023Leveraging Symmetries in Pick and Placecited by 18
  38. 2024Imagination Policy: Using Generative Point Cloud Models for Learning Manipulation Policiescited by 18

50 most cited of 59 All 59 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 2.

6 papers.

  1. 2025Two by Two: Learning Multi-Task Pairwise Objects Assembly for Generalizable Robot Manipulationcited by 12
  2. 2025AnyPlace: Learning Generalized Object Placement for Robot Manipulationcited by 10
  3. 2025A Practical Guide for Incorporating Symmetry in Diffusion Policycited by 4
  4. 2025OMP: One-step Meanflow Policy with Directional Alignmentcited by 4
  5. 2026Action Map Policy: Learning 3D Closed-loop Manipulation via Pixel Classificationcited by 0
  6. 2026Pix2Act: Image-Space Manipulation Policies with Equivariant Augmentationcited by 0