Research map / 3D vision / 3D generation
AdaDexGrasp: Adaptive Dexterous Grasping via 3D Visuo-Tactile Representation Fusion
Xirui Liang, Jingkai Xu, Jiaqi Liang, Yuran Wang, Ruochong Li, Yuanpei Chen, Masayoshi TOMIZUKA, Wei Zhan, Ruihai Wu
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 |
|---|---|---|---|
| add adaptation module (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 59% -> 83% | Improved |
| add adaptation module (ablation) | AdaDexGrasp · Sim Unseen Objects | grasp success rate: 78% -> 82% | Improved |
| add contact ID in grasp generation (ablation) | AdaDexGrasp · Sim Seen Objects | grasp success rate: 79% -> 91% | Improved |
| add contact ID in adaptation (ablation) | AdaDexGrasp · Sim Seen Objects | grasp success rate: 84% -> 91% | Improved |
| add hand point cloud (combined hand-object input) (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 61% -> 83% | Improved |
| add tactile features (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 67% -> 83% | Improved |
| use point clouds instead of images (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 64% -> 83% | Improved |
| add intermediate contact map (ablation) | AdaDexGrasp · Sim Seen Objects | grasp success rate: 82% -> 91% | Improved |
| add intermediate contact map (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 77% -> 83% | Improved |
| use combined SE(3)+joint proximity metric for adaptation pairs (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 61% -> 83% | Improved |
| use combined SE(3)+joint proximity metric for adaptation pairs (ablation) | AdaDexGrasp · Sim Unseen Categories | grasp success rate: 57% -> 83% | Improved |
| use tactile-intensity vector instead of dense force-field (ablation) | AdaDexGrasp · Sim small-data (50% data) | grasp success rate: 81% -> 88% | Improved |
Closest papers by meaning
- Touch in the Wild: Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper (NeurIPS 2025 · similarity 0.89)
- Multi-fingered Hand Grasps with Visuo-Tactile Fusion via Multi-Agent Deep Reinforcement Learning (AAAI 2025 · similarity 0.89)
- Contact Map Transfer with Conditional Diffusion Model for Generalizable Dexterous Grasp Generation (NeurIPS 2025 · similarity 0.89)
- Taccel: Scaling Up Vision-based Tactile Robotics via High-performance GPU Simulation (NeurIPS 2025 · similarity 0.88)
- Enhancing Tactile-based Reinforcement Learning for Robotic Control (NeurIPS 2025 · similarity 0.88)
- UniGraspTransformer: Simplified Policy Distillation for Scalable Dexterous Robotic Grasping (CVPR 2025 · similarity 0.88)
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.
