Research map / Detection and segmentation / CNN features
Predicting the Driver's Focus of Attention: The DR(eye)VE Project
Andrea Palazzi, Davide Abati, Simone Calderara, Francesco Solera, Rita Cucchiara
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add optical-flow branch to image branch (ablation) | multi-branch · DR(eye)VE test sequences | D_KL: 1.415 -> 1.399 | Improved |
| add optical-flow branch to image branch (ablation) | multi-branch · DR(eye)VE acting subsequences | D_KL: 1.826 -> 1.799 | Improved |
| add semantic-segmentation branch to image+flow (ablation) | multi-branch · DR(eye)VE test sequences | D_KL: 1.399 -> 1.398 | No clear effect |
| add semantic-segmentation branch to image+flow (ablation) | multi-branch · DR(eye)VE test sequences | IG: 0.033 -> 0.038 | Improved |
| add semantic branch to image branch (ablation) | multi-branch · DR(eye)VE test sequences | D_KL: 1.415 -> 1.413 | No clear effect |
| add image branch to flow+semantic (ablation) | multi-branch · DR(eye)VE test sequences | D_KL: 1.571 -> 1.398 | Improved |
| add image branch to flow branch (ablation) | multi-branch · DR(eye)VE test sequences | CC: 0.516 -> 0.558 | Improved |
| use multi-branch model instead of previous RGB-only model | Palazzi et al. · DR(eye)VE test sequences | D_KL: 1.48 -> 1.40 | Improved |
| use multi-branch model instead of previous model | Palazzi et al. · DR(eye)VE acting subsequences | IG: 0.20 -> 0.51 | Improved |
| use video-based multi-branch model instead of image-only MLNet | MLNet · DR(eye)VE test sequences | D_KL: 2.00 -> 1.40 | Improved |
| use task-trained multi-branch model instead of mean fixation baseline | Baseline Mean · DR(eye)VE acting subsequences | CC: 0.22 -> 0.41 | Improved |
| evaluate in rainy rather than sunny weather | Baseline Mean · DR(eye)VE test, weather | D_KL: 1.75 -> 1.53 | Improved |
Closest papers by meaning
- FIFA: Fine-grained Inter-frame Attention for Driver's Video Gaze Estimation (CVPR 2025 · similarity 0.88)
- SalM²: An Extremely Lightweight Saliency Mamba Model for Real-Time Cognitive Awareness of Driver Attention (AAAI 2025 · similarity 0.87)
- What Do You See in Vehicle? Comprehensive Vision Solution for In-Vehicle Gaze Estimation (CVPR 2024 · similarity 0.87)
- Leveraging Driver Field-of-View for Multimodal Ego-Trajectory Prediction (ICLR 2025 · similarity 0.86)
- On-Road Object Importance Estimation: A New Dataset and A Model with Multi-Fold Top-Down Guidance (NeurIPS 2024 · similarity 0.85)
- DriveGazen: Event-Based Driving Status Recognition Using Conventional Camera (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
