Introduction

The Honda Scenes dataset is a large scale annotated dataset created to enable dynamic scene classification. The dataset contains 80 hours of diverse high quality driving video data clips collected in the San Francisco Bay area. The dataset includes temporal annotations for road places, road environment, weather, and road surface conditions. HSD has the following specifications:

• The dataset contains 11 classes of road places - 3-way intersection, 4-way intersection, 5-way intersection, overhead bridge, railway crossing, construction zone, merge with gore on left, merge with gore on right, branch with gore on left, branch with gore on right, zebra crossing.
• The dataset spans 4 classes of road environments - rural, urban, highway and ramp and 4 weather conditions - rainy, sunny, cloudy, and foggy.
• Most classes have 3 temporal sub-classes , including approaching, entering, and passing.    The merge and branch classes have approaching and passing sub-classes, while the zebra-crossing class has just approaching.
• The dataset contains a total of about 20,000 instances spanned over these 11 classes.

The figures below outline key statistics in the dataset.

Dataset

This dataset corresponds to the paper, "Dynamic Traffic Scene Classification with Space-Time Coherence", as it appears in the proceedings of International Conference on Robotics and Automation (ICRA) 2019. In the current release, the dataset is only being made available to researchers in universities.

Download the dataset

The dataset is available for non-commercial usage. You must be affiliated with a university and use your university email address to make the request. Use this link to make the download request.

Citation

Please cite the following paper if you find the dataset useful in your work:

@article{narayanan2019dynamic,
  title={Dynamic Traffic Scene Classification with Space-Time Coherence},
  author={Narayanan, Athma and Dwivedi, Isht and Dariush, Behzad},
  journal={arXiv preprint arXiv:1905.12708}, 
  year={2019}
}