mlsanityData to 14 Oct 2026

Shape from Shading for Robotic Manipulation

arXiv 2304.11824 ↗First posted 24 Apr 2023cs.RO cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Controlling illumination can generate high quality information about object surface normals and depth discontinuities at a low computational cost. In this work we demonstrate a robot workspace-scaled controlled illumination approach that generates high quality information for table top scale objects for robotic manipulation. With our low angle of incidence directional illumination approach, we can precisely capture surface normals and depth discontinuities of monochromatic Lambertian objects. We show that this approach to shape estimation is 1) valuable for general purpose grasping with a single point vacuum gripper, 2) can measure the deformation of known objects, and 3) can estimate pose of known objects and track unknown objects in the robot's workspace.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 22 of its 53 references that are in the map. First posted on arXiv 24 Apr 2023.

Builds on

22 of 53 references.

References · 22

  1. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,613
  2. 2018QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulationcited by 743
  3. 2020NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesiscited by 496
  4. 2020NeRD: Neural Reflectance Decomposition from Image Collectionscited by 449
  5. 2020INeRF: Inverting Neural Radiance Fields for Pose Estimationcited by 382
  6. 2019A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithmscited by 142
  7. 2016Revisiting Active Perceptioncited by 94
  8. 2022NeRF-Supervision: Learning Dense Object Descriptors from Neural Radiance Fieldscited by 86
  9. 2017Intel RealSense Stereoscopic Depth Camerascited by 83
  10. 2021Unified Shape and SVBRDF Recovery using Differentiable Monte Carlo Renderingcited by 71
  11. 2022PS-NeRF: Neural Inverse Rendering for Multi-view Photometric Stereocited by 69
  12. 2018GelSlim: A High-Resolution, Compact, Robust, and Calibrated Tactile-sensing Fingercited by 52
10 more
  1. 2022S3-NeRF: Neural Reflectance Field from Shading and Shadow under a Single Viewpointcited by 30
  2. 2022PIE-Net: Photometric Invariant Edge Guided Network for Intrinsic Image Decompositioncited by 26
  3. 2022VIRDO: Visio-tactile Implicit Representations of Deformable Objectscited by 25
  4. 2023WildLight: In-the-wild Inverse Rendering with a Flashlightcited by 21
  5. 2017CNN based Learning using Reflection and Retinex Models for Intrinsic Image Decompositioncited by 19
  6. 2020Intrinsic Image Decomposition using Paradigmscited by 17
  7. 2022VIRDO++: Real-World, Visuo-tactile Dynamics and Perception of Deformable Objectscited by 13
  8. 2013From Shading to Local Shapecited by 12
  9. 2022Towards Learning Neural Representations from Shadowscited by 12
  10. 2022Using Collocated Vision and Tactile Sensors for Visual Servoing and Localizationcited by 6

Built on by

Not cited yet.