Verb Recognition on Ego4D
28.93Top-1 AccX-MIC
Evaluation results
| Method | Links | ||
|---|---|---|---|
| X-MICTemporal Attention=true, Trained on=Ego4D2024.03 | 28.93 | 27.65 | |
| X-MIC+DINOTemporal Attention=true, Trained on=Ego4D, Visual Encoder=CLIP + DINO2024.03 | 28.27 | 28.86 | |
| CLIP-Adapter*Temporal Attention=true, Trained on=Ego4D2024.03 | 27.32 | 24.54 | |
| A5Temporal Attention=true, Trained on=Ego4D2024.03 | 26.31 | 24.41 | |
| CLIP-AdapterTemporal Attention=false, Trained on=Ego4D2024.03 | 22.82 | 21.28 | |
| Vita-CLIPTemporal Attention=true, Trained on=Ego4D2024.03 | 22.66 | 24.13 | |
| CoOpTemporal Attention=false, Trained on=Ego4D2024.03 | 22.57 | 21.44 | |
| Co-CoOpTemporal Attention=false, Trained on=Ego4D2024.03 | 21.31 | 16.14 | |
| X-MICTemporal Attention=true, Trained on=Epic-Kitchens2024.03 | 18.1 | 26.58 | |
| A5Temporal Attention=true, Trained on=Epic-Kitchens2024.03 | 17.93 | 25.81 | |
| CLIP-AdapterTemporal Attention=false, Trained on=Epic-Kitchens2024.03 | 16.09 | 22.37 | |
| X-MIC+DINOTemporal Attention=true, Trained on=Epic-Kitchens, Visual Encoder=CLIP + DINO2024.03 | 16.01 | 24.6 | |
| CLIP-Adapter*Temporal Attention=true, Trained on=Epic-Kitchens2024.03 | 15.52 | 23.54 | |
| Co-CoOpTemporal Attention=false, Trained on=Epic-Kitchens2024.03 | 14.17 | 20.03 | |
| Vita-CLIPTemporal Attention=true, Trained on=Epic-Kitchens2024.03 | 13.47 | 21.11 | |
| CoOpTemporal Attention=false, Trained on=Epic-Kitchens2024.03 | 13.35 | 18.64 | |
| ZS CLIPTemporal Attention=false, Trained on=Zero-shot2024.03 | 2.18 | 2.88 |