US: Telenav, a leading provider of connected-car and location-based platform services, announced the world’s first open-source machine-learning technology designed to detect navigation features in street-level imagery. Telenav’s AI-based, map-making technology automatically detects navigation features using street-level imagery, which can reduce the effort required to create and enhance map data by as much as 10x. Telenav has crowdsourced more than 130-million street images via its OpenStreetCam™ mobile application and data collection continues to grow rapidly.
The company’s open-source AI initiative empowers developers to enhance the AI algorithms, and ensures that the next era of map making remains open and accessible to everyone.
Telenav has open sourced two key components of its broader AI-mapping platform:
- Machine Learning Algorithm – Used to recognize dozens of different types of road elements, including road signs and traffic lights; other items, such as lane and road width, will be added soon.
- Training sets – Consisting of over 50,000 tagged street-view images.
“Until now, all AI-based map-making technologies have been proprietary,” stated Philipp Kandal, Senior Vice President, Engineering for Telenav. “The openness of OSM is what originally made it a success, and now, with open AI technology, we can exponentially improve the map with the contribution of a large, passionate and talented community of developers.”
The company is also launching its “Telenav MapAI Contest” to encourage contributions to the open-source map-making platform. It invites interested participants to view details of the contest at https://competitions.codalab.org/competitions/19024. In addition to a cash prize of $10,000, the contributions to the map-making platform will be available for all mappers to use to improve OSM.