This is a curated collection of peer-reviewed papers related to attention and driving published in top transportation, human factors, and robotics venues since 2010.

The collection includes behavioral studies and applications where drivers' gaze allocation is explicitly measured (e.g. via an eye-tracker) or is used in some relevant practical application (e.g. driver assistance).

The following papers were excluded:

1. studies using modes of transportation other than cars (e.g. bicycles, motorcycles, trucks, buses, trains);
2. studies that rely only on indirect methods to assess drivers' attention (e.g. ego-vehicle sensor information);
3. studies that focused on drivers with medical issues or under the influence of alcohol or drugs;
4. uncited papers over 5 years old;
5. non-peer-reviewed papers (e.g. arXiv). An exception was made for reports from government organizations (e.g. NHTSA).

Papers in the collection are grouped into behavioral, application (grouped into 5 categories), and datasets. For each behavioral paper we provide link to paper, citation in bibtex format and tags. For the application papers we provide link to paper, link to code (if available), citation, and information on what dataset was used (private if data was unpublished or link(s) to public dataset(s)). For the dataset papers we provide a link to the paper where it was introduced, citation, link to the data, and a short summary of the data and annotations.

• Behavioral papers - psychological studies of drivers' attention
• Application papers - modeling or detection of drivers' attention for assistive and autonomous driving applications:
  • Modeling scene gaze - models that predict where the driver might look in the traffic scene
  • Detecting in-vehicle gaze - models for detecting what area inside the vehicle the driver is looking at
  • Detecting driver distraction - models for detecting manual, visual, and cognitive distractions
  • Detecting driver drowsiness - models for detecting driver drowsiness and fatigue
  • Driver action anticipation - models that anticipate drivers' maneuvers using gaze and vehicle information
  • Driver awareness estimation - models that project drivers' gaze onto objects in the traffic scene
  • Attention for self-driving - models that use attention mechanisms for automated driving
  • Papers with code - a list of papers across all applications that have public code available
• Datasets - a list of datasets with attention-related annotations

If you notice any errors or missing papers and code, please post an issue on this github.

If you used this repository in your research, please cite:

@article{kotseruba2022practical,
  title={Attention for Vision-Based Assistive and Automated Driving: A Review of Algorithms and Datasets},
  author={Kotseruba, Iuliia and Tsotsos, John K},
  journal={IEEE Transactions on Intelligent Transportation Systems},
  volume = {23},
  number = {11},
  pages = {19907--19928},
  year={2022}
}

@article{kotseruba2021behavioral,
  title={Behavioral Research and Practical Models of Drivers' Attention},
  author={Kotseruba, Iuliia and Tsotsos, John K},
  journal={arXiv preprint arXiv:2104.05677},
  year={2021}
}

This work is inspired by the database of papers on vision-based action prediction created by Amir Rasouli.