VERM: Leveraging Foundation Models to Create a Virtual Eye for Efficient 3D Robotic Manipulation
About
When performing 3D manipulation tasks, robots have to execute action planning based on perceptions from multiple fixed cameras. The multi-camera setup introduces substantial redundancy and irrelevant information, which increases computational costs and forces the model to spend extra training time extracting crucial task-relevant details. To filter out redundant information and accurately extract task-relevant features, we propose the VERM (Virtual Eye for Robotic Manipulation) method, leveraging the knowledge in foundation models to imagine a virtual task-adaptive view from the constructed 3D point cloud, which efficiently captures necessary information and mitigates occlusion. To facilitate 3D action planning and fine-grained manipulation, we further design a depth-aware module and a dynamic coarse-to-fine procedure. Extensive experimental results on both simulation benchmark RLBench and real-world evaluations demonstrate the effectiveness of our method, surpassing previous state-of-the-art methods while achieving 1.89x speedup in training time and 1.54x speedup in inference speed. More results can be found on our project website at https://verm-ral.github.io .
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Multi-task Robotic Manipulation | RLBench | Avg Success Rate83.6 | 16 | |
| Close Drawer | Real-world Evaluation 1.0 (unseen object placements) | Success Rate100 | 4 | |
| Flip Cup | Real-world Evaluation unseen object placements 1.0 | Success Rate80 | 4 | |
| Multi-task Robotic Manipulation | Real-world Evaluation unseen object placements 1.0 | Success Rate80 | 4 | |
| Open cabinet | Real-world Evaluation unseen object placements 1.0 | Success Rate80 | 4 | |
| Press sanitizer | Real-world Evaluation unseen object placements 1.0 | Success Rate90 | 4 | |
| Put block in bowl | Real-world Evaluation 1.0 (unseen object placements) | Success Rate0.8 | 4 | |
| Put object in drawer | Real-world Evaluation unseen object placements 1.0 | Success Rate0.7 | 4 | |
| Put object in shelf | Real-world Evaluation unseen object placements 1.0 | Success Rate80 | 4 | |
| stack blocks | Real-world Evaluation unseen object placements 1.0 | Success Rate0.8 | 4 |