Affectivahas developed an emotion recognition engine thanks to deep learning technology. It has spent years building an emotion recognition engine used for market research and being developed for use in cars at the moment. Emotion-aware cars may employ alerts to make distracted drivers give attention, relax drivers who are angry or find out how best to hand control back over to a human driver (for a highly-automated car).

Your future car might know if you are happy or sad

Affectiva’s system makes use of a camera focused on the driver’s faces, which identifies 33 facial landmarks, that information gets deduced by the computer’s neural network, based on the company’s deep learning of more than 5 million facial expressions. The computer recognizes 7 emotions, like fear, joy, surprise etc. in addition to other attributes including engagement.

The company has fed its network facial expressions from over 70 countries. It is working with a number of German and Japanese automakers at the moment. It is expected that a prototype using the technology will be there in 2018.