What is Mapillary?
Mapillary is an open source platform for sharing and using crowdsourced street-level imagery. It combines high-resolution 360-degree panoramic photos taken by users all over the world with computer vision algorithms to map and understand what is captured in those images.
Users can explore places virtually through immersive panoramic photo spheres mapped to OpenStreetMap. The imagery on Mapillary is captured by everything from phones and action cameras to cars equipped with specialized camera rigs. Over 1.5 million kilometers of imagery from 650,000 contributors make up the ever-growing openly licensed Mapillary Vistas photo dataset.
The platform uses computer vision and machine learning to automatically detect and outline objects like cars, people, traffic signs, and road markings in these images. This enables powerful visual search, analytics, augmented reality experiences, and training better algorithms. Everything from improving maps to inventorying street infrastructure and assets is possible by tapping into Mapillary's image database.
As an open platform, Mapillary is extensible to third-party services. Developers can build custom applications on top of the crowdsourced imagery through the Mapillary API and SDKs. Overall, Mapillary facilitates more informed decisions about the places people live, work, and travel with an open visual map of the world.