What's In a Smile? New AI Identity
|University of Bradford Researchers|
Your Smile, Your Identity What's in a smile? Beauty, kindness, friendship and now a lot of tech identifiers. Scientists at the University of Bradford in the UK have developed a computer based artificial intelligence system that maps the dynamics of a smile through facial muscle movements. Out of the box, it has 86% accuracy in identifying if the smile is that of a male or female. That's a first, but the potential is way beyond that.
Next G Biometrics: Tech Dynamics of Smiling
There are 49 landmarks on the face around the eyes, mouth, and down the nose when someone smiles. An algorithm has been created and tested by the University of Bradford team on 109 smiling faces. What's fascinating is that because the system measures underlying muscle movements of the face during a smile, the dynamic remains the same even if some physical features change following surgery. That means it's a potentially big biometric identifier.
Future of Smiles
The UK team believes this kind of facial recognition may be Next G. It's not dependent on 1 feature but on a dynamic that's unique to a person and very difficult to mimic or duplicate. By the way, researchers found female smiles are much more expressive and expansive. This is new biometric tech on the cutting edge. And, they believe the accuracy rate can be easily increased.