mlsanityData to 7 Oct 2026

SPLIT: SE(3)-diffusion via Local Geometry-based Score Prediction for 3D Scene-to-Pose-Set Matching Problems

arXiv 2411.10049 ↗First posted 15 Nov 2024cs.ROcited by 0

On the map: Diffusion models → Diffusion and flow policies for robotics

Abstract (arXiv)

To enable versatile robot manipulation, robots must detect task-relevant poses for different purposes from raw scenes. Currently, many perception algorithms are designed for specific purposes, which limits the flexibility of the perception module. We present a general problem formulation called 3D scene-to-pose-set matching, which directly matches the corresponding poses from the scene without relying on task-specific heuristics. To address this, we introduce SPLIT, an SE(3)-diffusion model for generating pose samples from a scene. The model's efficiency comes from predicting scores based on local geometry with respect to the sample pose. Moreover, leveraging the conditioned generation capability of diffusion models, we demonstrate that SPLIT can generate the multi-purpose poses, required to complete both the mug reorientation and hanging manipulation within a single model.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models. Builds on 15 of its 36 references that are in the map. First posted on arXiv 15 Nov 2024.

Builds on

15 of 36 references.

Closest ancestors · 1

Same direction, newest first

  1. 2023Diffusion-EDFs: Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation6 Sep

Other references · 14

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  3. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  4. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  5. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  6. 2020Convolutional Occupancy Networkscited by 760
  7. 2017Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metricscited by 259
  8. 2018Dense Object Nets: Learning Dense Visual Object Descriptors By and For Robotic Manipulationcited by 140
  9. 2018A micro Lie theory for state estimation in roboticscited by 117
  10. 2021Synergies Between Affordance and Geometry: 6-DoF Grasp Detection via Implicit Representationscited by 107
  11. 2021Volumetric Grasping Network: Real-time 6 DOF Grasp Detection in Cluttercited by 94
  12. 2022Equivariant Descriptor Fields: SE(3)-Equivariant Energy-Based Models for End-to-End Visual Robotic Manipulation Learningcited by 54
2 more
  1. 2024Deep Learning-Based Object Pose Estimation: A Comprehensive Surveycited by 39
  2. 2024Constrained 6-DoF Grasp Generation on Complex Shapes for Improved Dual-Arm Manipulationcited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023ReorientDiff: Diffusion Model based Reorientation for Object Manipulation
  2. 20236-DoF Stability Field via Diffusion Models
  3. 2026HeRO: Hierarchical 3D Semantic Representation for Pose-aware Object Manipulation
  4. 2024SPOT: SE(3) Pose Trajectory Diffusion for Object-Centric Manipulation
  5. 2023Act3D: 3D Feature Field Transformers for Multi-Task Robotic Manipulation
  6. 2025ISS Policy : Scalable Diffusion Policy with Implicit Scene Supervision
  7. 2025RoboTransfer: Controllable Geometry-Consistent Video Diffusion for Manipulation Policy Transfer
  8. 2023Diffusion-based Generation, Optimization, and Planning in 3D Scenes