Video Action Recognition on SS v2 (val)
8.9Top-1 Acc (K=2)EZ-CLIP
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| EZ-CLIPBackbone=ViT-142023.12 | 8.9 | 10.1 | 11.1 | 15.2 | |
| EZ-CLIPBackbone=ViT-162023.12 | 6.8 | 8.7 | 9.6 | 13.2 | |
| ViFi-CLIP2023.12 | 6.2 | 7.4 | 8.5 | 12.4 | |
| EZ-CLIPBackbone=ViT-322023.12 | 5.8 | 6.7 | 8 | 12.1 | |
| A52023.12 | 4.4 | 5.1 | 6.1 | 9.7 | |
| ActionCLIP2023.12 | 4.1 | 5.8 | 8.4 | 11.1 | |
| X-CLIP2023.12 | 3.9 | 4.5 | 6.8 | 10 | |
| Vanilla CLIP2023.12 | 2.7 | 2.7 | 2.7 | 2.7 |