Research map / Reinforcement learning / Visual and representation RL
DexMV: Imitation Learning for Dexterous Manipulation from Human Videos
Yuzhe Qin, Yueh-Hua Wu, Shaowei Liu, Hanwen Jiang, Ruihan Yang, Yang Fu, Xiaolong Wang
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 demonstrations (DAPG) to RL | TRPO · Relocate mustard bottle | success rate: 0.06 -> 0.93 | Improved |
| add demonstrations (SOIL) to RL | TRPO · Relocate sugar box | success rate: 0.00 -> 0.67 | Improved |
| use DAPG instead of SOIL | imitation learning · Relocate sugar box | success rate: 0.67 -> 0.00 | Hurt |
| use DAPG state-action demos instead of SOIL state-only | imitation learning · Pour | success (%): 3.5 -> 27.2 | Improved |
| use DAPG instead of SOIL | imitation learning · Place Inside | inside score: 27.9 -> 31.3 | Improved |
| increase demonstrations from 10 to 100 (ablation) | SOIL · Relocate tomato soup can | success rate: 0.36 -> 1.00 | Improved |
| increase demonstrations from 50 to 100 (ablation) | SOIL · Relocate tomato soup can | success rate: 0.59 -> 1.00 | Improved |
| use proposed retargeting instead of fingertip mapping (ablation) | SOIL · Relocate tomato soup can | success rate | Improved |
| add L2 smoothness norms to retargeting objective (ablation) | SOIL · Relocate tomato soup can | success rate | Improved |
| use two cameras with post-processing for hand pose (ablation) | SOIL · Relocate tomato soup can | success (%): 66.7 -> 93.3 | Improved |
| use dual camera instead of single camera (ablation) | SOIL · Relocate tomato soup can | success (%): 66.7 -> 84.7 | Improved |
| scale object size and friction in sim (ablation) | SOIL · Relocate tomato soup can | success rate | No clear effect |
Closest papers by meaning
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.90)
- Scaffolding Dexterous Manipulation with Vision-Language Models (NeurIPS 2025 · similarity 0.90)
- DexFlyWheel: A Scalable and Self-improving Data Generation Framework for Dexterous Manipulation (NeurIPS 2025 · similarity 0.89)
- ManiVideo: Generating Hand-Object Manipulation Video with Dexterous and Generalizable Grasping (CVPR 2025 · similarity 0.88)
- CyberDemo: Augmenting Simulated Human Demonstration for Real-World Dexterous Manipulation (CVPR 2024 · similarity 0.88)
- Cross-Embodiment Dexterous Grasping with Reinforcement Learning (ICLR 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.
