DR(eye)VE

a dataset for attention-based tasks
with applications to autonomous and assisted driving.

About the dataset

Composed by 74 video sequences of 5 mins each, we have captured and annotated more than 500,000 frames. The labeling contains drivers’ gaze fixations and their temporal integration providing task-specific saliency maps. Geo-referenced locations, driving speed and course complete the set of released data.

Features and diversity

Download

To download the dataset, we require an email address where we'll send a download link to. This will help us to keep in touch in case errors are found or as updates become available.

Email address Organization Get link

We believe in open research and we are happy if you find this data useful. If you use it, please cite our work.

 @article{dreyeve2018,
  title={Predicting the Driver's Focus of Attention: the DR(eye)VE Project},
  author={Palazzi, Andrea and Abati, Davide and Calderara, Simone and Solera, Francesco and Cucchiara, Rita},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2018}
  }
            

Code

The code related to the journal paper can be found at https://github.com/ndrplz/dreyeve.

Acknowledgments

We are really grateful to Federico Bolelli, Andrea Corbelli, Marcella Cornia, Matilde Loscalzo and Marco Venturelli for their valuable help in recording many of the driving sequences of this dataset.

Preview

Below you can find a preview of each one of the 74 video clips. Together with the GIFs, we included categories for each video (landascape, weather, daytime) and the driver.