Open-Vocabulary Semantic Segmentation on Replica OpenLex3D
45S ScoreHOV-SG
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| HOV-SGInput=ground-truth poses and depth2026.05 | 45 | 0 | 5 | 27 | 7 | 16 | |
| OpenMask3DInput=ground-truth poses and depth2026.05 | 43 | 1 | 7 | 29 | 10 | 10 | |
| LEXI-SGInput=ground-truth poses and depth2026.05 | 42 | 0 | 7 | 19 | 16 | 16 | |
| ConceptGraphsInput=ground-truth poses and depth2026.05 | 41 | 1 | 11 | 24 | 2 | 22 | |
| LEXI-SGInput=ground-truth poses and depth2026.05 | 41 | 1 | 3 | 14 | 16 | 26 | |
| HOV-SGInput=ground-truth poses and depth2026.05 | 33 | 2 | 4 | 18 | 8 | 36 | |
| OpenMask3DInput=ground-truth poses and depth2026.05 | 31 | 1 | 3 | 13 | 26 | 26 | |
| ConceptGraphsInput=ground-truth poses and depth2026.05 | 27 | 2 | 3 | 12 | 8 | 47 |