Data Set Street images for self-driving cars

Published on May 11th, 2017 | by Matthew Beckwith


Making it Easier to Train Autonomous Vehicles

Mapillary, a Swedish technology firm, has published the Mapillary Vistas Dataset consisting of 25,000 annotated street-level images to help researchers train autonomous driving systems. The images, taken at different times of the day and in varying weather conditions, come from 190 countries and Mapillary groups them into 100 different categories, including “bike lane,” “pedestrian area,” and “traffic island.” The Mapillary Vistas Dataset is free for both academic and commercial research.

Get the data.

Tags: , ,

About the Author

Matthew Beckwith

Matthew Beckwith is a policy fellow at the Center for Data Innovation. Previously, Matt worked as an analyst in the White House Office of Presidential Correspondence. Prior to that, he worked as an intern at the White House, as well as a speechwriting and policy intern on the O'Malley for President 2016 campaign. He has a Bachelor of Arts degree in Political Science from Boston College.

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to Top ↑

Show Buttons
Hide Buttons