3D Open Vocabulary Instance Segmentation On 2
Metrics
AP50 Base B6/N6
AP50 Base B8/N4
AP50 Novel B6/N6
AP50 Novel B8/N4
Results
Performance results of various models on this benchmark
Model Name | AP50 Base B6/N6 | AP50 Base B8/N4 | AP50 Novel B6/N6 | AP50 Novel B8/N4 | Paper Title | Repository |
|---|---|---|---|---|---|---|
| PLA | 46.9 | 59.0 | 9.8 | 8.6 | PLA: Language-Driven Open-Vocabulary 3D Scene Understanding | |
| OpenIns3D | - | - | 33.0 | 37.0 | OpenIns3D: Snap and Lookup for 3D Open-vocabulary Instance Segmentation | |
| Open3DIS | 50.0 | 60.8 | 29.0 | 26.3 | Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance | |
| Lowis3D | 51.8 | 58.7 | 15.8 | 13.8 | Lowis3D: Language-Driven Open-World Instance-Level 3D Scene Understanding | - |
0 of 4 row(s) selected.