Action Recognition on Epic-100 (test)
54.4Action AccuracyAVION
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| AVIONBackbone=ViT-L, Pretrain Data=WIT+Ego4D2023.09 | 54.4 | 73 | 65.4 | — | — | — | — | |
| LaViLaBackbone=TSF-L, Pretrain Data=WIT+Ego4D2023.09 | 51 | 72 | 62.9 | — | — | — | — | |
| MTVPretrain Data=WTS-60M2023.09 | 50.5 | 69.9 | 63.9 | — | — | — | — | |
| OmnivoreBackbone=Swin-B, Pretrain Data=IN-(21k+1k)+K400+SUN2023.09 | 49.9 | 69.5 | 61.7 | — | — | — | — | |
| AVIONBackbone=ViT-B, Pretrain Data=WIT+Ego4D2023.09 | 49.1 | 70 | 59.8 | — | — | — | — | |
| MeMViTPretrain Data=K6002023.09 | 48.4 | 71.4 | 60.3 | — | — | — | — | |
| MoViNetPretrain Data=N/A2023.09 | 47.7 | 72.2 | 57.3 | — | — | — | — | |
| LaViLaBackbone=TSF-B, Pretrain Data=WIT+Ego4D2023.09 | 46.9 | 69 | 58.4 | — | — | — | — | |
| X-ViT#frames=16x2021.06 | 44.3 | 68.7 | 56.4 | — | — | — | — | |
| ViViT-L/16x22021.06 | 44 | 66.4 | 56.8 | — | — | — | — | |
| ViViT-LPretrain Data=IN-21k+K4002023.09 | 44 | 66.4 | 56.8 | — | — | — | — | |
| TimeMambalearning_mode=fine-tuned, sampled_frames=162024.03 | 42.8 | 66.6 | 53.3 | 63.2 | — | — | — | |
| X-ViT#frames=8x2021.06 | 41.5 | 66.7 | 53.3 | — | — | — | — | |
| TimeSformerlearning_mode=fine-tuned, sampled_frames=162024.03 | 41.3 | 63.8 | 52.4 | 60.4 | — | — | — | |
| IPLBackbone=I3D, Pretrain Data=K4002023.09 | 41 | 68.6 | 51.2 | — | — | — | — | |
| SlowFast2021.06 | 38.5 | 65.6 | 50 | — | — | — | — | |
| TSM2021.06 | 38.3 | 67.9 | 49 | — | — | — | — | |
| TBN2021.06 | 36.7 | 66 | 47.2 | — | — | — | — | |
| TRN2021.06 | 35.3 | 65.9 | 45.4 | — | — | — | — | |
| TSN2021.06 | 33.2 | 60.2 | 46 | — | — | — | — | |
| AVIONEnsemble=true2024.06 | — | — | — | — | 71.7 | 65.8 | 54.3 | |
| CA2ST2025.03 | — | — | — | — | 46.7 | 28.8 | 17.4 | |
| CAST2025.03 | — | — | — | — | 47.1 | 30.3 | 18.4 | |
| CAVA2025.03 | — | — | — | — | 35.7 | 22.3 | 10.6 | |
| EgoVideo-VEnsemble=false2024.06 | — | — | — | — | 72.9 | 68.7 | 56.2 | |
| EgoVideo-VEnsemble=true2024.06 | — | — | — | — | 73.1 | 69.8 | 56.8 | |
| MotionFormer-HR2025.03 | — | — | — | — | 35.7 | 20 | 10.2 | |
| Omnivore2025.03 | — | — | — | — | 28.7 | 17.1 | 8.7 | |
| TIM2025.03 | — | — | — | — | 51.3 | 36.1 | 23.4 | |
| VideoMAE-L2025.03 | — | — | — | — | 47.5 | 29.4 | 17.9 |