| visual-place-recognition-on-sf-xl-night | SAGE ViT-L (without cross-image encoder) | #1 | Recall@1: 0.599Recall@10: 0.779 |
| visual-place-recognition-on-amstertime | SAGE ViT-L (without cross-image encoder) | #2 | Recall@1: 65.6Recall@10: 90.4Recall@5: 86.7 |
| visual-place-recognition-on-sf-xl-test-v1 | SAGE ViT-L (without cross-image encoder) | #2 | Recall@1: 95.7Recall@10: 97.7Recall@5: 97.4 |
| visual-place-recognition-on-svox-snow | SAGE ViT-B (without cross-image encoder) | #2 | Recall@1: 99.3 |
| visual-place-recognition-on-svox-snow | SAGE ViT-L (without cross-image encoder) | #9 | Recall@5: 99.7Recall@10: 99.9000 |
| visual-place-recognition-on-pittsburgh-250k-test | SAGE ViT-L (without cross-image encoder) | #3 | Recall@1: 96.9Recall@10: 99.6Recall@5: 99.2 |
| visual-place-recognition-on-pittsburgh-30k-test | SAGE ViT-L (without cross-image encoder) | #3 | Recall@1: 94.7Recall@5: 97.7Recall@10: 98.4 |
| visual-place-recognition-on-sf-xl-occlusion | SAGE ViT-B (without cross-image encoder) | #3 | Recall@1: 0.579 |
| visual-place-recognition-on-svox-overcast | SAGE ViT-L (without cross-image encoder) | #3 | Recall@1: 98.4000Recall@5: 99.5Recall@10: 99.9000 |
| visual-place-recognition-on-svox-sun | SAGE ViT-L (without cross-image encoder) | #3 | Recall@1: 98.2Recall@5: 99.5Recall@10: 99.8 |
| visual-place-recognition-on-eynsham | SAGE ViT-L (without cross-image encoder) | #4 | Recall@1: 92.3Recall@10: 96.7Recall@5: 95.9 |
| visual-place-recognition-on-nordland | SAGE ViT-L (without cross-image encoder) | #4 | Recall@1: 94.8Recall@5: 98.2Recall@10: 98.9 |
| visual-place-recognition-on-svox-night | SAGE ViT-L (without cross-image encoder) | #5 | Recall@1: 97.3Recall@5: 99.4 |
| visual-place-recognition-on-svox-night | SAGE ViT-B (without cross-image encoder) | #9 | Recall@10: 99.8 |
| visual-place-recognition-on-sf-xl-test-v2 | SAGE ViT-L (without cross-image encoder) | #6 | Recall@1: 94.8Recall@5: 98.3Recall@10: 99 |
| visual-place-recognition-on-mapillary-val | SAGE ViT-L (without cross-image encoder) | #7 | Recall@1: 94.2Recall@5: 97.8Recall@10: 98.1 |
| visual-place-recognition-on-tokyo247 | SAGE ViT-L (without cross-image encoder) | #7 | Recall@1: 98.1 |
| visual-place-recognition-on-sped | SAGE ViT-L (without cross-image encoder) | #8 | Recall@1: 92.1Recall@5: 96.2Recall@10: 96.9 |