Mapillary is powered by world-class computer vision
3D reconstruction and semantic segmentation connect images and extract map data at scale for smart cities, geospatial services, and autonomous vehicles.
Use our technology
You can process imagery for your own services or get training data for your algorithms.
All Mapillary imagery is anonymized for privacy compliance. Our deep learning based algorithms are optimized for blurring faces and license plates on street-level images.
Mapillary uses Structure from Motion to find the relative positions of images and create smooth transitions between them. Our OpenSfM
pipeline is available as open source.
Each image is pixel-wise labeled with 97 object classes. We have the best-performing algorithm for semantic segmentation of road imagery, based on Cityscapes and Mapillary Vistas benchmarks.
Traffic sign recognition
We detect more than 900 types of traffic signs in 67 countries. When the same sign appears in multiple images, its position in space is triangulated and the traffic sign is extracted as a map feature.
Training data for segmentation algorithms
25,000 manually annotated images, pixel-wise segmented into 97 classes. Covering 6 continents and different weather, season, time of day, camera, and viewpoint, Mapillary Vistas help you train algorithms that recognize objects in road scenes across the world.
Machine-labeled training data
Find training data for your specific needs in Mapillary's global image repository. Millions of images from street scenes around the world are automatically segmented into 97 object classes with pixel-wise object labels.
"The Mapillary Vistas help improve the robustness of our models and verify their performance in applications on markets all over the world."
Let’s talk about what you can do with our processing services and training data.
We believe that there is a need for an independent and neutral provider of street-level imagery and map data, untied to any particular mapping platform. We are committed to building a global service for everyone.