Video Action Recognition on Kinetics-GYM
90.8Top-1 AccuracySMILE
Evaluation Results
| Method | Links | |
|---|---|---|
| SMILEEvaluation Protocol=Finetuning, Intermediate Dataset=K400 [35]2025.04 | 90.8 | |
| AIMEvaluation Protocol=Finetuning, Intermediate Dataset=None2025.04 | 90.3 | |
| UMTEvaluation Protocol=Finetuning, Intermediate Dataset=K700 [8]2025.04 | 89.9 | |
| ViCLIPEvaluation Protocol=Finetuning, Intermediate Dataset=Intervid-10M [84]2025.04 | 89.7 | |
| CLIPEvaluation Protocol=Finetuning, Intermediate Dataset=None2025.04 | 88 | |
| ILAEvaluation Protocol=Finetuning, Intermediate Dataset=None2025.04 | 82.7 | |
| ViFi-CLIPEvaluation Protocol=Finetuning, Intermediate Dataset=None2025.04 | 81.5 | |
| X-CLIPEvaluation Protocol=Finetuning, Intermediate Dataset=None2025.04 | 75.2 | |
| SMILEEvaluation Protocol=Linear Probing, Intermediate Dataset=K400 [35]2025.04 | 30.2 | |
| ViCLIPEvaluation Protocol=Linear Probing, Intermediate Dataset=Intervid-10M [84]2025.04 | 27.3 | |
| UMTEvaluation Protocol=Linear Probing, Intermediate Dataset=K700 [8]2025.04 | 26.4 | |
| CLIPEvaluation Protocol=Linear Probing, Intermediate Dataset=None2025.04 | 20.7 |