Context-aware Robotic Grasping on Real-world context-aware grasping scenes
98Object SuccessPoint2Act
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Point2ActInput=RGB2025.08 | 98 | 93 | 73 | 16.5 | |
| Point2ActInput=RGB-D2025.08 | 96 | 92 | 69 | 9.5 | |
| MLLM*Input=RGB-D2025.08 | 89 | 68 | 41 | 3.5 | |
| LERF-TOGOInput=RGB2025.08 | 87 | 45 | 28 | 102.5 | |
| GraspMolmoInput=RGB-D2025.08 | 78 | 59 | 36 | 3.5 | |
| F3RMInput=RGB2025.08 | 75 | 41 | 25 | 22.5 | |
| GraspSplatsInput=RGB-D2025.08 | 73 | 40 | 27 | 70 | |
| GaussianGrasperInput=RGB-D2025.08 | 62 | 26 | 15 | 150 |