Scale AI, a startup based in San Francisco, the University of Waterloo, and the University of Toronto have released a dataset containing 56,000 images and 7,000 point clouds from LiDAR to advance the ability of autonomous vehicles to operate in wintry weather. The data comes from eight cameras that captured 20 kilometers of driving in harsh conditions in Ontario and can help autonomous vehicles learn to drive when snow obscures lane markings and their sensors. The data includes labels for objects such as cars, bicycles, and pedestrians.
Image: Elmuzzerino