Image Based Localization On Cvusa 1
Métriques
Recall@1
Recall@10
Recall@5
Recall@top1%
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Recall@1 | Recall@10 | Recall@5 | Recall@top1% |
|---|---|---|---|---|
| university-1652-a-multi-view-multi-source | 43.91 | 74.58 | 66.38 | 91.78 |
| cv-cities-advancing-cross-view-geo | 99.19 | 99.85 | 99.80 | 99.92 |
| joint-representation-learning-and-keypoint | 52.50 | - | - | 96.52 |
| transgeo-transformer-is-all-you-need-for | 94.08 | 99.04 | 98.36 | 99.77 |
| sample4geo-hard-negative-sampling-for-cross | 98.68 | 99.78 | 99.68 | 99.87 |
| beyond-geo-localization-fine-grained | 95.43 | 99.34 | 98.86 | 99.86 |
| each-part-matters-local-patterns-facilitate | 85.79 | 96.98 | 95.38 | 99.41 |
| simple-effective-and-general-a-new-backbone | 96.34 | 99.50 | 99.10 | 99.86 |