Facebook’s technology is getting even better. The social media giant has an experimental algorithm that can recognize people even when it is difficult to see their faces. The technology identifies people by other characteristics like body shape, hairstyle or clothing choices.
A look at Facebook’s new technology
There are plenty of similar technologies found on social network websites and in churches and retail stores. Facebook’s artificial intelligence lead, Yann LeCun, wants to have technology that can view humans even when the face isn’t seen well. LeCun explained:
“There are a lot of cues we use. People have characteristic aspects, even if you look at them from the back. “For example, you can recognise Mark Zuckerberg very easily, because he always wears a gray T-shirt.”
As part of the project, the research team pulled together nearly 40,000 public photos from Flickr. Some of the photos showed full faces clearly, and other photos show faces turned away from the camera. They ran the photos through a neural network. The last algorithm could identify people’s faces with an 83% accuracy rate, which is quite impressive.
Facebook shows off the new technology
The project was presented during this year’s Computer Vision and Pattern Recognition event. Facebook hopes the algorithm could give better power to photo apps such as Facebook Moments. The latter is a new app that allows people to privately send photos to friends and get photos they didn’t send. Moments also sorts through the photos stored on the phone to separate them via events. It also tags Facebook friends who are recognized in the photos. It revolutionizes social photo sharing.
The downside to all this technology is that it raises serious concerns for privacy advocates. Facial recognition could benefit companies and businesses that seek new ways to market products, but it also violates consumer privacy.