Cognitive Distraction Detection on BDD-A (Leave-one-dataset-out)
65.24AccuracyEyeCue
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| EyeCueBackbone=TimeSformerK600, Training-free=false2026.05 | 65.24 | 71 | |
| Heatmap-basedBackbone=SVM, Training-free=false2026.05 | 61.95 | 57 | |
| EyeCueBackbone=VideoMAEK400, Training-free=false2026.05 | 60.95 | 59 | |
| TimeSformerK400Backbone=ViT-B/16, Training-free=false2026.05 | 60.71 | 62 | |
| TimeSformerK600Backbone=ViT-B/16, Training-free=false2026.05 | 60.52 | 67 | |
| GazeVQABackbone=CLIP ViT-B/32, Training-free=false2026.05 | 59.15 | 49 | |
| Voila-ABackbone=CLIP ViT-L/14, Training-free=false2026.05 | 58.81 | 57 | |
| EgoVideoBackbone=ViT-B/14, Training-free=false2026.05 | 58.57 | 65 | |
| VideoMAEK400Backbone=ViT-B/16, Training-free=false2026.05 | 57.38 | 62 | |
| VideoLLaMA3Backbone=ViT-B/16, Training-free=false2026.05 | 54.86 | 47 | |
| Video-LLaVABackbone=OpenCLIP-L/14, Training-free=false2026.05 | 52.43 | 37 | |
| GazeGPTBackbone=GLM-4.5V-AWQ, Training-free=true2026.05 | 51 | 47 | |
| InternVideo2s1-1BBackbone=ViT-B/14, Training-free=false2026.05 | 50.16 | 45 | |
| GazeLLMBackbone=Gemini-2.5-Pro, Training-free=true2026.05 | 44.53 | 2 |