Facial ID as plain as the smile on your face
Capture those laughter lines
The way you smile could uniquely identify you, and provide a basis for new facial recognition technology.
Physicists at Stony Brook University in New York have developed a facial recognition technique, which relies on the pattern of muscle movement beneath the skin, rather than matching surface characteristics.
Traditional techniques can easily be fooled, according to E Guan, who heads the Stony Brook team. They rely on measuring distance between the eyes, and the eyes and mouth, for example, which means simple makeup or dark glasses can throw off the system.
However, the new method compares images of a person, taken fractions of a second apart, while they are in the process of smiling. Guan then analyses how the skin around the mouth moves between shots by tracking the change position and direction of tiny wrinkles in the skin.It doesn't actually need a full smile, though, and is sensitive enough to produce a map of features - even when people were trying to keep a straight face.
The system successfully identified the four lab members in images that had fooled other methods, according to Nature, The team is now testing it on 30 new pictures.
The research may also help identify certain nerve disorders by detecting facial asymmetries. ®