Action Detection On Ucf Sports
Métriques
Frame-mAP 0.5
Video-mAP 0.2
Video-mAP 0.5
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Frame-mAP 0.5 | Video-mAP 0.2 | Video-mAP 0.5 |
|---|---|---|---|
| multi-region-two-stream-r-cnn-for-action | 84.52 | 94.83 | 94.67 |
| multi-region-two-stream-r-cnn-for-action | 82.30 | 94.82 | 94.82 |
| finding-action-tubes | 68.1 | - | 75.8 |
| dance-with-flow-two-in-one-stream-action | - | - | 96.52 |
| dance-with-flow-two-in-one-stream-action | - | - | 92.74 |
| tube-convolutional-neural-network-t-cnn-for | 86.7 | - | - |
| finding-action-tubes-with-a-sparse-to-dense | - | 94.3 | 93.8 |