Facebook can now detect your presence in photos without needing to see your face.

This is thanks to a new algorithm developed in Facebook’s artificial intelligence labs that uses aspects like hair styles, body types, clothes and even poses to determine who’s in a photo with 83 per cent accuracy.

Facial recognition technology is already quite far advanced, but according to New Scientist Facebook’s head of artificial intelligence, Yann LeCun, wanted to see if it was possible to use other non-facial cues to identify people when their faces weren’t clearly captured. At 83% accuracy, the answer is clearly “yes”.

The technology promises to be very useful for features like tagging friends in a collection of photos, and sorting them according to who was present without manually going through the whole lot.

Of course, such advancements also lead inevitably to questions of privacy. While New Scientist says this sort of algorithm could be useful to people who want to be alerted every time their image appears on the internet for purposes of control and privacy, it could just as easily be used to identify people who wish to keep their identities to themselves.

It’s a hot topic in today’s age, where technology and privacy often clash, and unfortunately privacy advocates and industry pundits can’t seem to agree on a middle path.

But at least the debate is taking place, and it is hopefully just be a matter of time before common ground is found.

What do you think? Does this sort of technological advancement worry you, or are you of the opinion that the notion of privacy is a very 20th-century concept? As always, let us know…

[Source – New Scientist, Image – CC BY-SA 2.0]