Cross-Modal Scene Retrieval (Image to Point Cloud) on 3RScan
14.01Scene Matching Recall@1CrossOver
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| CrossOverModality Pair=I → P2025.02 | 14.01 | 49.04 | 66.88 | 83.44 | 12.77 | 36.17 | 70.21 | |
| Inst. BaselineModality Pair=I → P2025.02 | 8.92 | 30.57 | 43.31 | 64.33 | 0.04 | 19.15 | 42.55 | |
| ULIP-2Modality Pair=I → P2025.02 | 1.27 | 5.1 | 7.01 | 12.74 | 0.04 | 4.26 | 12.77 | |
| PointBindModality Pair=I → P2025.02 | 1.27 | 4.46 | 9.55 | 17.2 | 2.13 | 4.26 | 8.51 |