| Paper | Code | Frame-mAP 0.5 | Video-mAP 0.2 | Video-mAP 0.5 | ModelName | ReleaseDate |
|---|---|---|---|---|---|---|
| Tube Convolutional Neural Network (T-CNN) for Action Detection in Videos | ✓ Link | 86.7 | T-CNN | 2017-03-30 | ||
| Multi-region two-stream R-CNN for action detection | 84.52 | 94.83 | 94.67 | MR-TS R-CNN | 2016-09-17 | |
| Multi-region two-stream R-CNN for action detection | 82.30 | 94.82 | 94.82 | TS R-CNN | 2016-09-17 | |
| Finding Action Tubes | ✓ Link | 68.1 | 75.8 | Action Tubes | 2014-11-21 | |
| Finding Action Tubes with a Sparse-to-Dense Framework | 94.3 | 93.8 | DTS | 2020-08-30 | ||
| Dance with Flow: Two-in-One Stream Action Detection | ✓ Link | 96.52 | Two-in-one Two Stream | 2019-04-01 | ||
| Dance with Flow: Two-in-One Stream Action Detection | ✓ Link | 92.74 | Two-in-one | 2019-04-01 |