Now Everyone Can Help Self-driving Cars Learn the Ropes...Er, Roads

It's no secret that self-driving cars need massive training data sets to recognize and categorize objects. Enter crowdsourcing. Swedish startup Mapillary has crowdsourced 260 million images to create “the largest and most diverse dataset” of its kind. With recent additional funding of $15M, Mapillary plans to expand its platform and offer more automotive specific integration. 

We noticed they gamefied image contributions with Leaderboards and wonder how well this scales. It's one thing to get people to submit cat videos, but quite another to crowdsource fairly ho-hum street-level images. 

Ps. We want to meet the guy who contributed +53k photos and personally buy him a cat.


Want to receive more content like this in your inbox?