Detection of Figure 4 on Seizure Video Dataset
92.2AccuracyQwen
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| QwenModel Architecture=Qwen, Signal Enhancement (Pose)=false2026.05 | 92.2 | 60 | 37.5 | 46.2 | |
| ViViTModel Architecture=ViViT, Signal Enhancement (Pose)=false2026.05 | 78.9 | 25.6 | 70 | 33.2 | |
| InternModel Architecture=Intern, Signal Enhancement (Pose)=false2026.05 | 78.9 | 21.1 | 50 | 29.6 | |
| Pose+QwenModel Architecture=Qwen, Signal Enhancement (Pose)=true2026.05 | 56.8 | 13.5 | 62.5 | 22.2 | |
| CNNModel Architecture=CNN, Signal Enhancement (Pose)=false2026.05 | 51.1 | 8.6 | 56.7 | 12.6 | |
| Pose+InternModel Architecture=Intern, Signal Enhancement (Pose)=true2026.05 | 29.8 | 11.9 | 100 | 21.3 |