Action Recognition on EK100, SSV2, and K400
71.6Overall Harmonic MeanCAST
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CASTGFLOPs/View=3912023.11 | 71.6 | — | |
| CASTGFLOPs/View=3912025.03 | 71.6 | 77.9 | |
| OMNIVORE2023.11 | 70.8 | — | |
| OMNIVORE2025.03 | 70.8 | 77.2 | |
| MTV-HRGFLOPs/View=9302023.11 | 69.8 | — | |
| MTV-HRGFLOPs/View=9302025.03 | 69.8 | 74.8 | |
| ST-AdapterGFLOPs/View=6072023.11 | 67.3 | — | |
| MFormerGFLOPs/View=11852023.11 | 67.3 | — | |
| ST-AdapterGFLOPs/View=6072025.03 | 67.3 | 75.5 | |
| MFormer-HRGFLOPs/View=11852025.03 | 67.3 | 73.7 | |
| AIMGFLOPs/View=4042023.11 | 66.7 | — | |
| AIMGFLOPs/View=404, Parameter-efficient tuning with adapters=true2025.03 | 66.7 | 75.4 | |
| ViViT FEGFLOPs/View=9902023.11 | 66.6 | — | |
| VideoMAEGFLOPs/View=1802023.11 | 66.6 | — | |
| VIVIT FEGFLOPs/View=9902025.03 | 66.6 | 73 | |
| VideoMAE-BGFLOPs/View=180, Parameter-efficient tuning with adapters=true2025.03 | 66.6 | 75.8 | |
| CLIPGFLOPs/View=1402023.11 | 56.5 | — | |
| CLIPGFLOPs/View=140, Parameter-efficient tuning with adapters=true2025.03 | 56.5 | 59.5 | |
| BEVTGFLOPs/View=2822025.03 | — | 75.3 | |
| EVLGFLOPs/View=5922025.03 | — | 71.2 | |
| MVIT-BGFLOPs/View=1702025.03 | — | 73.4 | |
| TimeSformer-BGFLOPs/View=23802025.03 | — | 70.4 | |
| Video Swin-LGFLOPs/View=2822025.03 | — | 75.8 |