The data sets we use is from Udacity self-driving-car public data source.

We use the data set includes driving in Mountain View California and neighboring cities during daylight conditions. It contains over 65,000 labels across 9,423 frames collected from Point Grey research cameras running at full resolution of 1920x1200 at 2 hz. The data set was annotated by CrowdAI using a combination of machine learning and humans. Besides, there are 3 labels for the data set, which are Car, Truck and Pedestrian. A CSV file involved in the data source, which is used to record the location and dimension of bounding boxes of respective objects for each image, the bounding boxes are denoted as the labels, xmin, ymin, xmax, ymax.

Data Source Link:

https://github.com/udacity/self-driving-car/tree/master/annotations.

Previous: State-of-the-art Next: Methodology