Video Recognition on UCF-101 (Base/Novel/HM Metrics)
96.1Base ScoreOST
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| OSTMode=Tuning2024.11 | 96.1 | 68.9 | 80.3 | |
| Zero-ViCLIP + MSTA + LccMode=Efficient transfer, GFLOPs=179.842024.11 | 96 | 72.9 | 82.9 | |
| Zero-ViCLIP + MSTAMode=Efficient transfer, GFLOPs=179.842024.11 | 95.6 | 72.8 | 82.7 | |
| ViCLIPMode=Tuning, GFLOPs=1612024.11 | 95.1 | 71.5 | 81.6 | |
| Zero-ViCLIP + LORAMode=Efficient transfer, GFLOPs=179.802024.11 | 95.1 | 72.3 | 82.1 | |
| Zero-ViCLIP + AdaptFormerMode=Efficient transfer, GFLOPs=179.802024.11 | 94.9 | 72.7 | 82.3 | |
| ViFi-CLIPMode=Tuning, GFLOPs=2812024.11 | 92.5 | 68.5 | 78.7 | |
| Vanilla CLIPMode=Tuning, GFLOPs=2812024.11 | 92.3 | 67.9 | 78.2 | |
| A5Mode=Adapting, GFLOPs=2842024.11 | 90.9 | 40.8 | 56.3 | |
| XCLIPMode=Adapting, GFLOPs=1452024.11 | 90.7 | 59.5 | 71.9 | |
| ActionCLIPMode=Adapting, GFLOPs=2822024.11 | 90.5 | 58.7 | 71.2 | |
| Zero-ViCLIPMode=Efficient transfer, GFLOPs=161.742024.11 | 81.7 | 69.8 | 75.3 | |
| Vanilla CLIPMode=Zero-shot, GFLOPs=2812024.11 | 78 | 63.8 | 70.2 |