Oral automatisms recognition on Clinical seizure videos
53.3AccuracyIntern
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| InternModel Type=VLM, Signal Enhancement=None2026.05 | 53.3 | 35.9 | 45.2 | 40 | |
| QwenModel Type=VLM, Signal Enhancement=None2026.05 | 50 | 35.4 | 54.8 | 43 | |
| CNNModel Type=Traditional2026.05 | 45.6 | 37.5 | 88.9 | 52.4 | |
| Crop+InternModel=Intern, Signal Enhancement=Crop2026.05 | 45.6 | 33.3 | 58.1 | 42.4 | |
| ViViTModel Type=Traditional2026.05 | 44.4 | 36.2 | 83.3 | 50.3 | |
| Crop+QwenModel=Qwen, Signal Enhancement=Crop2026.05 | 44.4 | 35.4 | 74.2 | 47.9 |