Cognitive Distraction Detection on DR(eye)VE (Leave-one-dataset-out)
60.2AccuracyEyeCue
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EyeCueBackbone=TimeSformerK600, Training-free=false2026.05 | 60.2 | 67 | |
| InternVideo2s1-1BBackbone=ViT-B/14, Training-free=false2026.05 | 59.31 | 58 | |
| VideoMAEK400Backbone=ViT-B/16, Training-free=false2026.05 | 58.85 | 64 | |
| EgoVideoBackbone=ViT-B/14, Training-free=false2026.05 | 55.65 | 56 | |
| TimeSformerK600Backbone=ViT-B/16, Training-free=false2026.05 | 55.41 | 61 | |
| TimeSformerK400Backbone=ViT-B/16, Training-free=false2026.05 | 55.4 | 55 | |
| Voila-ABackbone=CLIP ViT-L/14, Training-free=false2026.05 | 54.91 | 45 | |
| EyeCueBackbone=VideoMAEK400, Training-free=false2026.05 | 54.67 | 62 | |
| Video-LLaVABackbone=OpenCLIP-L/14, Training-free=false2026.05 | 53.75 | 51 | |
| GazeVQABackbone=CLIP ViT-B/32, Training-free=false2026.05 | 51.97 | 65 | |
| GazeGPTBackbone=GLM-4.5V-AWQ, Training-free=true2026.05 | 51.29 | 42 | |
| VideoLLaMA3Backbone=ViT-B/16, Training-free=false2026.05 | 51.17 | 39 | |
| Heatmap-basedBackbone=SVM, Training-free=false2026.05 | 49.39 | 14 | |
| GazeLLMBackbone=Gemini-2.5-Pro, Training-free=true2026.05 | 49.14 | 8 |