Long Video Activity Recognition On Breakfast
Métriques
mAP
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | mAP |
|---|---|
| actionvlad-learning-spatio-temporal | 60.20 |
| videograph-recognizing-minutes-long-human | 63.14 |
| graph-based-high-order-relation-modeling-for | 65.86 |
| adafocus-towards-end-to-end-weakly-supervised | 69.6 |
| adafocus-towards-end-to-end-weakly-supervised | 70.4 |
| adafocus-towards-end-to-end-weakly-supervised | 79.2 |
| adafocus-towards-end-to-end-weakly-supervised | 79.5 |
| timeception-for-complex-action-recognition | 61.82 |