Mapillary Inc.

Brooklyn, NY, United States


Mapillary is a new approach to street-level imagery. Our platform creates 3D reconstructions of locations from the street-level photos for extracting geospatial data using computer vision. Mapillary allows the capability for anyone to collect street level photos with smartphone apps and off-the-shelf equipment. With advanced features like time filtering and automated traffic sign detection, Mapillary is ideal for projects which require fast turnaround, frequent updates and fresh coverage.



Mapillary offers the following services: i) IMAGE SEGMENTATION: 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. ii) 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. iii) 3D RECONSTRUCTION: 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. iv) PRIVACY BLURRING : 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. v) MAPILLARY VISTAS: Training data for user specific needs in Mapillary's global image repository. 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.

Services Provided:

Application Development, Data Conversion/Migration, GIS Strategy and Planning, Hosting Services, System Integration