Visual Place Recognition On Pittsburgh 30K
Métriques
Recall@1
Recall@5
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Recall@1 | Recall@5 |
|---|---|---|
| towards-seamless-adaptation-of-pre-trained | 92.8 | 97.7 |
| patch-netvlad-multi-scale-fusion-of-locally | 88.7 | 94.5 |
| generalized-contrastive-optimization-of | 81.94 | 94.56 |
| progeo-generating-prompts-through-image-text | 93.0 | 98.3 |
| emerging-properties-in-self-supervised-vision | 70.13 | - |
| rethinking-visual-geo-localization-for-large | 90.45 | - |
| rethinking-visual-geo-localization-for-large | 90.4 | 95.7 |
| revisit-anything-visual-place-recognition-via | 93.1 | - |
| pair-vpr-place-aware-pre-training-and | 95.4 | 97.5 |
| eigenplaces-training-viewpoint-robust-models | 92.5 | - |
| dinov2-learning-robust-visual-features | 78.32 | - |
| mixvpr-feature-mixing-for-visual-place | 91.52 | 95.9 |
| boq-a-place-is-worth-a-bag-of-learnable | 93.7 | 97.1 |
| pair-vpr-place-aware-pre-training-and | 94.7 | 97.2 |
| effovpr-effective-foundation-model | 93.9 | 97.4 |
| netvlad-cnn-architecture-for-weakly | 86.08 | - |
| anyloc-towards-universal-visual-place | 87.66 | - |
| anyloc-towards-universal-visual-place | 54.97 | - |
| boq-a-place-is-worth-a-bag-of-learnable | 92.4 | - |