Berkeley DeepDrive Dataset

Explore 100,000 HD video sequences of over 1,100-hour driving experience across many different times in the day, weather conditions, and driving scenarios. The video sequences also include GPS locations, IMU data, and timestamps. White paper on this dataset - BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning -

Data and Resources

Additional Info

Field Value
Source University of California, Berkeley
Author Fisher Yu, Haofeng Chen, Xin Wang, Wenqi Xian, Yingying Chen, Fangchen Liu, Vashisht Madhavan, Trevor Darrell
Maintainer Fisher Yu
Associated Tasks Image Recognition, Image Parsing, Tracking
Geographical Area USA: New York, San Francisco, Berkeley, Bay Area
Number of Instances 100,000 video sequences
Package Description 100,000 video sequences with annotations for objects including bus, traffic light, traffic sign, person, bike, truck, motor, car, train, and rider.
Dataset has missing values False