Soon enough, Facebook may be able to recognize the faces of your friends, family and co-workers just as precisely as you can.
The social media giant is further developing its facial verification
technology to make it nearly as accurate as the human eye, according to a
new blog post from the company and a report from MIT Technology Review.
When asked if the faces of two unfamiliar people are the same, the
average human will answer correctly 97.53% of the time, MIT says.
Facebook's technology will be able to tell faces apart 97.25% of the
time, which the company says is 25% more accurate than it was before.
Facebook's new software - known as DeepFace - performs facial
verification, which distinguishes whether or not two images show the
same face. This is not to be confused with facial recognition, which
helps put a name to the face, although Facebook's Yaniv Taiman, who
works on the company's AI team, tells MIT that DeepFace may improve
facial recognition as well.
DeepFace accomplishes this in two steps.
First,
it corrects the subject's face so that the person is facing forward in
the image. It uses a 3D model of an "average" forward-looking face to
nail down this angle.
Then, the software uses a method known as 'Deep Learning', which
means it simulates a neural network that can create a numerical
description of reoriented face. If the software finds similar enough
facial descriptions for two different images, it concludes that they
must be the same face.
Facial verification isn't new to Facebook. In fact, the social network began suggesting friends in tagged photos back in 2010.
The DeepFace software, however, will likely prevent the website from
mistakenly tagging you in photos as a friend that may have similar
facial features.
DeepFace is just a research project for now, according to MIT, but
researchers will be presenting the technology at the IEEE Conference on
Computer Vision and Pattern Recognition in June.
No comments:
Post a Comment