Research map / 3D vision / Face and shape
Active 3D Shape Reconstruction from Vision and Touch
Edward J. Smith, David Meger, Luis Pineda, Roberto Calandra, Jitendra Malik, Adriana Romero, Michal Drozdzal
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use the proposed reconstruction model instead of the cited T_G model | T_G reconstruction · T_G, 0 grasps | Chamfer distance: 25.586 -> 24.864 | Improved |
| Use the proposed reconstruction model instead of the cited T_G model | T_G reconstruction · T_G, 1 grasp | Chamfer distance: 9.016 -> 8.220 | Improved |
| Use the proposed reconstruction model instead of the cited V&T_G model | V&T_G reconstruction · V&T_G, 0 grasps | Chamfer distance: 2.653 -> 2.538 | Improved |
| Use the proposed reconstruction model instead of the cited V&T_G model | V&T_G reconstruction · V&T_G, 1 grasp | Chamfer distance: 2.637 -> 2.486 | Improved |
| Select a dataset-wide fixed trajectory by lowest average error rather than choose actions randomly | Active touch policy · T_P, 5 actions | CD ratio after 5 actions / initial CD: 36.38 -> 29.85 | Improved |
| Select a dataset-wide fixed trajectory by lowest average error rather than choose actions randomly | Active touch policy · T_G, 5 actions | CD ratio after 5 actions / initial CD: 25.83 -> 23.04 | Improved |
| Use nearest-neighbor action selection rather than random actions | Active touch policy · V&T_P, 5 actions | CD ratio after 5 actions / initial CD: 94.56 -> 91.79 | Improved |
| Use nearest-neighbor action selection rather than random actions | Active touch policy · V&T_G, 5 actions | CD ratio after 5 actions / initial CD: 90.65 -> 88.53 | Improved |
| Choose the fixed trajectory by lowest average error rather than most common best action | Dataset-specific active touch policy · T_P, 5 actions | CD ratio after 5 actions / initial CD: 32.40 -> 29.85 | Improved |
| Choose the fixed trajectory by lowest average error rather than most common best action | Dataset-specific active touch policy · T_G, 5 actions | CD ratio after 5 actions / initial CD: 23.46 -> 23.04 | Improved |
| Choose the fixed trajectory by lowest average error rather than most common best action | Dataset-specific active touch policy · V&T_P, 5 actions | CD ratio after 5 actions / initial CD: 93.59 -> 92.36 | Improved |
| Choose the fixed trajectory by lowest average error rather than most common best action | Dataset-specific active touch policy · V&T_G, 5 actions | CD ratio after 5 actions / initial CD: 89.39 -> 89.31 | No clear effect |
Closest papers by meaning
- Touch2Shape: Touch-Conditioned 3D Diffusion for Shape Exploration and Reconstruction (CVPR 2025 · similarity 0.91)
- Tactile DreamFusion: Exploiting Tactile Sensing for 3D Generation (NeurIPS 2024 · similarity 0.87)
- Tactile MNIST: Benchmarking Active Tactile Perception (NeurIPS 2026 · similarity 0.87)
- AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors (ICLR 2025 · similarity 0.87)
- Tactile-Augmented Radiance Fields (CVPR 2024 · similarity 0.86)
- PhysTacGen: Physics-Aware Visual-Tactile Sensor Image Generation (NeurIPS 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
