ZeroDexGrasp: Zero-Shot Task-Oriented Dexterous Grasp Synthesis with Prompt-Based Multi-Stage Semantic Reasoning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917085639081984 |
|---|---|
| author | Jian, Juntao Wei, Yi-Lin Mou, Chengjie Lin, Yuhao Zhu, Xing Shen, Yujun Zheng, Wei-Shi Hu, Ruizhen |
| author_facet | Jian, Juntao Wei, Yi-Lin Mou, Chengjie Lin, Yuhao Zhu, Xing Shen, Yujun Zheng, Wei-Shi Hu, Ruizhen |
| contents | Task-oriented dexterous grasping holds broad application prospects in robotic manipulation and human-object interaction. However, most existing methods still struggle to generalize across diverse objects and task instructions, as they heavily rely on costly labeled data to ensure task-specific semantic alignment. In this study, we propose \textbf{ZeroDexGrasp}, a zero-shot task-oriented dexterous grasp synthesis framework integrating Multimodal Large Language Models with grasp refinement to generate human-like grasp poses that are well aligned with specific task objectives and object affordances. Specifically, ZeroDexGrasp employs prompt-based multi-stage semantic reasoning to infer initial grasp configurations and object contact information from task and object semantics, then exploits contact-guided grasp optimization to refine these poses for physical feasibility and task alignment. Experimental results demonstrate that ZeroDexGrasp enables high-quality zero-shot dexterous grasping on diverse unseen object categories and complex task requirements, advancing toward more generalizable and intelligent robotic grasping. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_13327 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ZeroDexGrasp: Zero-Shot Task-Oriented Dexterous Grasp Synthesis with Prompt-Based Multi-Stage Semantic Reasoning Jian, Juntao Wei, Yi-Lin Mou, Chengjie Lin, Yuhao Zhu, Xing Shen, Yujun Zheng, Wei-Shi Hu, Ruizhen Robotics Task-oriented dexterous grasping holds broad application prospects in robotic manipulation and human-object interaction. However, most existing methods still struggle to generalize across diverse objects and task instructions, as they heavily rely on costly labeled data to ensure task-specific semantic alignment. In this study, we propose \textbf{ZeroDexGrasp}, a zero-shot task-oriented dexterous grasp synthesis framework integrating Multimodal Large Language Models with grasp refinement to generate human-like grasp poses that are well aligned with specific task objectives and object affordances. Specifically, ZeroDexGrasp employs prompt-based multi-stage semantic reasoning to infer initial grasp configurations and object contact information from task and object semantics, then exploits contact-guided grasp optimization to refine these poses for physical feasibility and task alignment. Experimental results demonstrate that ZeroDexGrasp enables high-quality zero-shot dexterous grasping on diverse unseen object categories and complex task requirements, advancing toward more generalizable and intelligent robotic grasping. |
| title | ZeroDexGrasp: Zero-Shot Task-Oriented Dexterous Grasp Synthesis with Prompt-Based Multi-Stage Semantic Reasoning |
| topic | Robotics |
| url | https://arxiv.org/abs/2511.13327 |