Video on Toyota Smarthome dataset
Metric: CV2 (higher is better)
Metric: CV2 (higher is better)
| # | Model↕ | CV2▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | EPAM-Net | 67.8 | Yes | EPAM-Net: An Efficient Pose-driven Attention-gui... | 2024-08-10 | Code |
| 2 | UNIK | 65 | No | UNIK: A Unified Framework for Real-world Skeleto... | 2021-07-19 | Code |
| 3 | π-ViT | 64.8 | Yes | Just Add $π$! Pose Induced Video Transformers fo... | 2023-11-30 | Code |
| 4 | NPL | 54.6 | No | Recognizing Actions in Videos from Unseen Viewpo... | 2021-03-30 | - |
| 5 | VPN (RGB + Pose) | 53.5 | Yes | VPN: Learning Video-Pose Embedding for Activitie... | 2020-07-06 | Code |
| 6 | Separable STA (RGB + Pose) | 50.3 | Yes | - | - | - |
| 7 | I3D | 45.1 | Yes | Quo Vadis, Action Recognition? A New Model and t... | 2017-05-22 | Code |
| 8 | I3D + Non Local | 43.9 | Yes | Non-local Neural Networks | 2017-11-21 | Code |
| 9 | Dense Trajectories | 23.7 | No | Improved Dense Trajectory with Cross Streams | 2016-04-29 | - |