3D Instance Segmentation on Replica 8 scenes
24.7mAPSOLE
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SOLEInputs=3D only, Region Proposal Network: Name=Not required (proposal-free), Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=Yes2026.04 | 24.7 | 31.8 | 40.3 | — | — | |
| SpaCeFormerInputs=3D only, Region Proposal Network: Name=Not required (proposal-free), Region Proposal Network: No GT=Yes, Region Proposal Network: Open vocab=Yes2026.04 | 24.1 | 31.8 | 37.1 | 0.14 | — | |
| BoxOVISInputs=3D + 2D, Region Proposal Network: Name=Box detector + SAM, Region Proposal Network: No GT=Yes, Region Proposal Network: Open vocab=Yes2026.04 | 24 | 31.8 | 37.4 | 43.7 | 312 | |
| OpenTrack3DInputs=3D + 2D, Region Proposal Network: Name=YOLO-World + SAM2, Region Proposal Network: No GT=Yes, Region Proposal Network: Open vocab=Yes2026.04 | 23.9 | 36.4 | 47.6 | 62 | 443 | |
| Open-YOLO 3DInputs=3D + 2D, Region Proposal Network: Name=Mask3D (2023), Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 23.7 | 28.6 | 34.8 | 16.6 | 119 | |
| Details MatterInputs=3D + 2D, Region Proposal Network: Name=Mask3D + GSAM, Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 22.6 | 31.7 | 37.7 | 597 | 4,264 | |
| Details Matter (3D)Inputs=3D, Region Proposal Network: Name=Mask3D + GSAM, Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 22 | 26.7 | 32.5 | — | — | |
| Details Matter (2D)Inputs=2D, Region Proposal Network: Name=Mask3D + GSAM, Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 20.8 | 32.4 | 38.5 | — | — | |
| PoVoInputs=3D + 2D, Region Proposal Network: Name=SAM (2023), Region Proposal Network: No GT=Yes, Region Proposal Network: Open vocab=Yes2026.04 | 20.8 | 28.7 | 34.4 | — | — | |
| Open3DISInputs=3D + 2D, Region Proposal Network: Name=ISBNet (2023) + SAM, Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 18.5 | 24.5 | 28.2 | 188 | 1,343 | |
| Open3DIS (3D prop. only)Inputs=3D, Region Proposal Network: Name=ISBNet (2023), Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 14.9 | 18.8 | 23.6 | 35.1 | 251 | |
| OV-MAPInputs=3D + 2D, Region Proposal Network: Name=SAM (2023), Region Proposal Network: No GT=Yes, Region Proposal Network: Open vocab=Yes2026.04 | 14.2 | 19.6 | 28.1 | — | — | |
| OpenMask3DInputs=3D + 2D, Region Proposal Network: Name=Mask3D (2023), Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 13.1 | 18.4 | 24.2 | 547.3 | 3,909 | |
| OVIR-3DInputs=3D + 2D, Region Proposal Network: Name=Detic (2022), Region Proposal Network: No GT=Yes, Region Proposal Network: Open vocab=Yes2026.04 | 11.1 | 20.5 | 27.5 | 52.7 | 376 | |
| OpenScene-2DInputs=3D + 2D, Region Proposal Network: Name=Mask3D (2023), Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 10.9 | 15.6 | 17.3 | 317.3 | 2,267 | |
| OpenScene-3DInputs=3D, Region Proposal Network: Name=Mask3D (2023), Region Proposal Network: No GT=No, Region Proposal Network: Open vocab=No2026.04 | 8.2 | 10.5 | 12.6 | 4.3 | 31 |