Facebook Can Now Recognize You In Photos Even If Your Face Is Covered
LatestYou know how you do that thing where you cover your face in pictures so even if your friends put them up on Facebook (without your permission, obviously) no one will be able to tell who you are? Turns out Facebook knows exactly who you are even without seeing your face.
New Scientist reports that Facebook’s got a new algorithm that allows recognition without needing to resort to anything as basic as facial features. Instead, the experimental (meaning it’s coming soon, don’t worry) algorithm looks at other recognizable characteristics to pick out who’s who in all those party pics that were shot in low light at bad angles.
From New Scientist:
“There are a lot of cues we use. People have characteristic aspects, even if you look at them from the back,” [Yann] LeCun says [head of Artificial Intelligence]. “For example, you can recognise Mark Zuckerberg very easily, because he always wears a gray T-shirt.”
The research team pulled almost 40,000 public photos from Flickr – some of people with their full face clearly visible, and others where they were turned away – and ran them through a sophisticated neural network.
The final algorithm was able to recognise individual people’s identities with 83 per cent accuracy. It was presented earlier this month at the Computer Vision and Pattern Recognition conference in Boston, Massachusetts.
While this could be a great feature for Facebook’s business (“moments” may be using this technology in the near future), it also raises some concerns about privacy. For instance, why does Facebook need to know who’s in every photo or what you’re doing? What if it starts tagging you without permission? Yes, it’s great that Facebook can alert you to photos of you being posted, but as New Scientist points out, the flipside of that coin may be a little bit more sinister.
“If, even when you hide your face, you can be successfully linked to your identify, that will certainly concern people,” says Ralph Gross at Carnegie Mellon University in Pittsburgh, Pennsylvania, who says the algorithm is impressive. “Now is a time when it’s important to discuss these questions.”
Contact the author at [email protected].
Image via Getty