Research map / Reinforcement learning / Visual and representation RL
DexH2R: A Benchmark for Dynamic Dexterous Grasping in Human-to-Robot Handover
Youzhuo Wang, Jiayi Ye, Chuyang Xiao, Yiming Zhong, Heng Tao, Hang Yu, Yumeng Liu, Jingyi Yu, Yuexin Ma
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 3D point-cloud observations (DP3) instead of RGB-D images (DP) | Diffusion Policy approaching motion · DexH2R Easy Mode | approach success rate: 39.4 -> 66.3 | Improved |
| use diffusion-policy with 3D point clouds instead of auto-regressive MotionNet | Approaching motion generation · DexH2R Easy Mode | safety rate: 25.9 -> 50.1 safe; 71.1 -> 66.3 succ | Mixed |
| tighten alignment threshold from 10cm to 5cm (Hard Mode) | MotionNet · DexH2R test | approach success rate: 71.1 -> 26.6 | Hurt |
| add pretraining on synthetic DexGraspNet before fine-tuning on DexH2R (ablation) | DexgraspAnything · DexH2R grasp generation | succ6: 50.62 -> 52.81 | Improved |
| add pretraining on synthetic DexGraspNet before fine-tuning on DexH2R (ablation) | DexgraspAnything · DexH2R grasp generation | succ1: 89.38 -> 88.75 | No clear effect |
| add pretraining on synthetic DexGraspNet before fine-tuning on DexH2R (ablation) | cVAE · DexH2R grasp generation | succ1: 76.56 -> 79.38 | Improved |
| add pretraining on synthetic DexGraspNet (ablation) | cVAE · DexH2R grasp generation | succ6: 35.56 -> 35.00 | No clear effect |
Closest papers by meaning
- ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning (CVPR 2025 · similarity 0.89)
- GenH2R: Learning Generalizable Human-to-Robot Handover via Scalable Simulation Demonstration and Imitation (CVPR 2024 · similarity 0.88)
- MobileH2R: Learning Generalizable Human to Mobile Robot Handover Exclusively from Scalable and Diverse Synthetic Data (CVPR 2025 · similarity 0.88)
- DexFlyWheel: A Scalable and Self-improving Data Generation Framework for Dexterous Manipulation (NeurIPS 2025 · similarity 0.87)
- DexGrasp Anything: Towards Universal Robotic Dexterous Grasping with Physics Awareness (CVPR 2025 · similarity 0.86)
- Grasp as You Say: Language-guided Dexterous Grasp Generation (NeurIPS 2024 · 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.
