Action Classification on Gym288
61.9Accuracy (Per-Class)TQN
Evaluation results
| Method | Links | ||
|---|---|---|---|
| TQNPretrained dataset=K400, Modality=RGB, # frames in training=dense2021.04 | 61.9 | 89.6 | |
| LT-S3DPretrained dataset=K400, Modality=RGB, # frames in training=dense2021.04 | 57.9 | 86.3 | |
| GST-50Pretrained dataset=ImageNet, Modality=RGB, # frames in training=82021.04 | 46.9 | 83.8 | |
| TSMPretrained dataset=ImageNet, Modality=two-stream, # frames in training=32021.04 | 46.5 | 83.1 | |
| TRNmsPretrained dataset=ImageNet, Modality=two-stream, # frames in training=32021.04 | 43.3 | 82 | |
| ST-S3DPretrained dataset=K400, Modality=RGB, # frames in training=82021.04 | 42.4 | 75.8 | |
| TSNPretrained dataset=ImageNet, Modality=two-stream, # frames in training=32021.04 | 37.6 | 79.9 | |
| I3DPretrained dataset=K400, Modality=RGB, # frames in training=82021.04 | 28.2 | 66.7 |