Text Based Person Retrieval On Icfg Pedes
Métriques
R@1
R@10
R@5
mAP
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | R@1 | R@10 | R@5 | mAP |
|---|---|---|---|---|
| rasa-relation-and-sensitivity-aware | 65.28 | 85.12 | 80.4 | 41.29 |
| plip-language-image-pre-training-for-person | 64.25 | 86.32 | 80.88 | - |
| towards-unified-text-based-person-retrieval-a | 68.51 | - | - | 41.22 |
| from-data-deluge-to-data-curation-a-filtering | 68.35 | 87.53 | 83.10 | 42.60 |
| mars-paying-more-attention-to-visual | 67.60 | 85.79 | 81.47 | 44.93 |
| vgsg-vision-guided-semantic-group-network-for | 63.05 | 84.36 | 78.43 | - |
| cross-modal-implicit-relation-reasoning-and | 63.46 | 85.82 | 80.25 | 38.06 |
| a-simple-and-robust-correlation-filtering | 57.18 | - | - | - |
| an-empirical-study-of-clip-for-text-based | 65.05 | 85.47 | 80.34 | 39.83 |
| noisy-correspondence-learning-for-text-to | 67.68 | 87.36 | 82.47 | 40.06 |
| semantically-self-aligned-network-for-text-to | 54.23 | - | - | - |
| cross-modal-adaptive-dual-association-for | - | - | - | 39.85 |