Text Based Person Retrieval On Rstpreid 1
Métriques
R@1
R@10
R@5
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | R@1 | R@10 | R@5 |
|---|---|---|---|
| dssl-deep-surroundings-person-separation | 39.05 | 73.95 | 62.60 |
| cross-modal-adaptive-dual-association-for | - | - | - |
| cross-modal-implicit-relation-reasoning-and | 60.20 | 81.30 | 88.20 |
| noisy-correspondence-learning-for-text-to | 65.35 | 89.90 | 83.95 |
| from-data-deluge-to-data-curation-a-filtering | 66.85 | 91.10 | 86.45 |
| mars-paying-more-attention-to-visual | 67.55 | 91.35 | 86.65 |
| towards-unified-text-based-person-retrieval-a | 67.50 | 91.45 | 85.70 |
| an-empirical-study-of-clip-for-text-based | 61.95 | 88.75 | 83.55 |
| rasa-relation-and-sensitivity-aware | 66.90 | 91.35 | 86.50 |