From Grasps to Dexterity: Large-Scale Grasp Pretraining for Dexterous Manipulation
About
Large-scale dexterous grasp datasets encode rich priors over hand-object interaction, but their use has largely been confined to grasp generation and pick-and-place manipulation. We study whether such data can instead support functional dexterity in articulated tool use, where a robot must acquire a tool, maintain contact, and operate its functional moving parts. We adapt a hierarchical imitation learning framework that combines high-level hand sub-goal prediction with a low-level goal-conditioned controller. We construct a 355k-trajectory grasp-pretraining dataset from large-scale dexterous grasp annotations and use it to pretrain the low-level controller. The controller is then fine-tuned on downstream task demonstrations. To evaluate this setting, we introduce DexCraft, a simulation benchmark with six articulated tool-use tasks requiring coordinated finger motion. Across simulation and real-world experiments, our approach outperforms end-to-end diffusion policy baselines and hierarchical policies trained from scratch. In the real world, it improves full-task success by 33.3 percentage points over DP3. These results show that grasp datasets can serve not only as resources for grasp synthesis, but also as scalable pretraining data for contact-rich dexterous manipulation. Videos are shown on https://yingyuan0414.github.io/grasp2dexterity/ .
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Dexterous Manipulation | DexCraft 20 demonstrations | SprayBot Success Rate59.2 | 6 | |
| Dexterous Manipulation | DexCraft 50 demonstrations | SprayBot Success Rate61.4 | 6 | |
| Dexterous Manipulation | DexCraft 100 demonstrations | SprayBot Success Rate68.4 | 6 | |
| Articulated Tool Use | Real-world Syringe 1.0 | Grasp Success Rate88.9 | 3 | |
| Articulated Tool Use | Real-world Spray Bottle 1.0 | Grasp Success Rate88.9 | 3 | |
| Articulated Tool Use | Real-world Scissors 1.0 | Grasp Success Rate66.7 | 3 | |
| Articulated Tool Use | Real-world Tools Average 1.0 | Grasp Success Rate77.8 | 3 |