3D Visual Grounding on ScanRefer ScanNet v2 (val)
93.4Unique AccMCLN
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| MCLNGT object segmentation=true, GT semantic labels=true2025.12 | 93.4 | 54.9 | 60.6 | |
| BUTD-DETRGT object segmentation=true, GT semantic labels=true2025.12 | 92.5 | 52.6 | 58.5 | |
| ViL3DRelGT object segmentation=true2025.12 | 92 | 51.8 | 59.6 | |
| PanoGrounderGT object segmentation=true2025.12 | 91.7 | 58.5 | 64.9 | |
| 3D-VisTAGT object segmentation=true2025.12 | 89.5 | 49.9 | 57.2 |