ZYT, DJI’s automotive company, has developed the AD4CHE dataset. Using drone-powered aerial surveys, ZYT collected data on highways and expressways in Chinese urban scenes to analyse driving congestion scenarios.
ZYT AD4CHE dataset
ZYT’s AD4CHE dataset, or Aerial Dataset for Chinese Highways and Expressways is a dataset made up of 552,960 frames across 53,761 roads covering a distance of 6,540.7km.
The dataset includes images across a range of city scenes where cars are travelling a maximum of 80km/h with heading angle, yaw rate, lateral offset, and other datapoints for each vehicle in the dataset.
ZYT states the dataset can be used for the analysis of traffic behaviour in various scenarios, providing a baseline for algorithms, and quickly generate autonomous driving scenarios for testing.
Access the dataset
ZYT has made the dataset available upon request and approval by it’s team. You can submit an application to get your hands on the dataset via ZYT’s website.
To find out more on the technicals of the dataset, you can get access to the AD4CHE format description document via this link.
Source: ZYT
