Canadian Adverse Driving Conditions Dataset Released!

Tuesday, February 4, 2020

WISE Lab led by prof. Czarnecki and TRAIL Lab led by prof. Waslander have released the Canadian Adverse Driving Conditions Dataset. The dataset is accompanied by a paper with the same title by Matthew Pitropov, Danson Garcia, Jason Rebello, Michael Smart, Carlos Wang, Krzysztof Czarnecki, Steven Waslander.

Dataset views: dynamic object bounding boxes shown in camera image and lidar pointcloud.

Camera view

CADCD camera view

Lidar view

CADCD lidar view

The release has attracted some media attention, including the WIRED Magazine article "Snow and Ice Pose a Vexing Obstacle for Self-Driving Cars" and CTV's video segment about the dataset release "Team working to train self-driving cars for winter driving". CBC published an article and audio interview "How Waterloo researchers are giving self-driving cars 'Canadian vision'". Global News also had a video segment and interview "Can self-driving cars handle Canadian winters?".

Scale.AI, a company whose technology was leveraged in creating the dataset, published a blog article "Open Sourcing the World's First AV Dataset for Wintry Environments".

Additionally, University of Waterloo Engineering published a story "Teaching self-driving cars how to handle the worst of winter" and University of Toronto Engineering published an article "Can self-driving cars handle a Canadian winter? We’re about to find out".

Remote video URL