Teach autonomous vehicles to understand their surroundings
Train visual perception models with rich datasets of street scenes and traffic signs from around the world. Scale up map updates to keep up with the needs of HD maps.
Robust and accurate visual perception models
Train perception models for autonomous driving with the Mapillary Vistas Dataset and the Mapillary Traffic Sign Dataset to develop vehicles that understand street scenes and recognize traffic signs all over the world in different seasons, weather and lighting conditions, and viewpoints.

Learn more about the datasets
Continuous map updates
License map data generated from hundreds of thousands of images that collaborators add to the Mapillary platform daily. Add imagery collected by commercial and autonomous fleets and share data between different endpoints so all vehicles will benefit from the data that the other ones are collecting.

Mapillary’s computer vision technology processes imagery fast and delivers automatically extracted map features such as traffic signs to help keep maps and navigation systems up to date.
Trusted by
Why Mapillary?
High diversity
Street-level imagery from more than 190 countries all over the world, different seasons, times of day, and vantage points—driving, cycling, or walking.
Fast map updates
Automatically extracted map data available from hundreds of thousands of images added to the database every day.
Training datasets
Highly diverse and detailed training data for models understanding whole street scenes or traffic signs in particular.
Support for imagery from a broad range of cameras, and a variety of tools for pulling map data into the applications and workflows you need.
Start accessing imagery and map data
Mapillary is the platform that makes street-level images and map data available to scale and automate mapping.
We’re committed to building a global service for everyone.