Language-grounded 3D semantic segmentation on GraspClutter6D
52.543D IoUClutt3R-Seg
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Clutt3R-SegCLIP model variant=Duoduo CLIP2026.02 | 52.54 | 47.2 | |
| Clutt3R-Seg-CLIP model variant=Standard CLIP2026.02 | 50.61 | 47.6 | |
| ConceptGraphs2026.02 | 32.59 | 148.5 | |
| GraspSplats2026.02 | 17.38 | 81.1 |