Object Detection
70,000/10,000/20,000 images for train/val/test, 1.8M objects.
100K driving videos (40 seconds each) collected from more than 50K rides, covering New York, San Francisco Bay Area, and other regions.
Contains diverse scene types such as city streets, residential areas, and highways. Recorded in diverse weather conditions at different times of the day.
10 taksk in total: Lane detection, object detection, semantic segmentation, instance segmentation, multi-object tracking, segmentation tracking and more.
70,000/10,000/20,000 images for train/val/test, 1.8M objects.
7,000/1,000/2,000 images for train/val/test, 120K masks.
1,400/200/400 videos for train/val/test, 160K instances, 4M objects.
154/32/37 videos for train/val/test, 25K instances, 480K masks.
7,000/1,000/2,000 images for train/val/test, 40 object classes.
70,000/10,000/20,000 images for train/val/test, 8 main categories.
70,000/10,000/20,000 images for train/val/test, 8 main categories.
6 weather conditions, 6 scene types, 3 distinct times of the day
GPS/IMU recordings with visual input and the driving trajectories.
Diverse weather, road and daytime conditions.
We provide documents and tools for inspection, preparation, and evaluation of the BDD100K dataset.
You can simply log in and download the data in your browser after agreeing to BDD100K license.
We provides scripts to parse and visualize the labels, and a tool to display the trajectories.
We use a consistent data annotation format across all different tasks. We choose the Scalabel [link] format for this.
We provide evaluation scripts, online testing servers and challenges to verify your algorithm.
BDD100K is compatible with the labels generated by Scalabel. The labels are released in Scalabel Format.