Action Recognition On H2O 2 Hands And Objects
Métriques
Actions Top-1
Hand Pose
Object Label
Object Pose
RGB
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Actions Top-1 | Hand Pose | Object Label | Object Pose | RGB |
|---|---|---|---|---|---|
| revisiting-skeleton-based-action-recognition | 83.47 | 2D | No | No | Yes |
| disentangling-and-unifying-graph-convolutions | 50.83 | 3D | No | No | No |
| sharp-segmentation-of-hands-and-arms-by-range | 91.73 | 3D | Yes | 2D | No |
| hierarchical-temporal-transformer-for-3d-hand | 86.36 | 3D (est.) | Yes (est.) | No | Yes |
| interactive-spatiotemporal-token-attention | 89.09 | 3D | No | Yes | No |
| h2o-two-hands-manipulating-objects-for-first | 79.25 | 3D | No | Yes | No |
| transformer-based-unified-recognition-of-two | 90.90 | 3D (est.) | Yes (est.) | Yes (est.) | Yes |
| slowfast-networks-for-video-recognition | 77.69 | No | No | No | Yes |
| spatial-temporal-graph-convolutional-networks-1 | 73.86 | 3D | No | Yes | No |
| in-my-perspective-in-my-hands-accurate | 91.32 | 2D | Yes | 2D | No |
| on-the-utility-of-3d-hand-poses-for-action | 93.39 | 3D | No | No | Yes |