Napa Sae-Bae from the Polytechnic Institute of New York University is working on a new method of authentication for touchscreen devices using hand gestures and finger swipes. This could provide a desirable login alternative to passwords on tablets and smartphones, recognizing individuals’ biological features to give them access.
The doctoral student at the department of computer science built an algorithm to detect unique biological traits from hand shape, finger movement, and finger length. She created an iPad app that matched people to the gestures they made on the touchscreen, like rotating their open palm or opening a clenched fist against the screen. In experiments with 34 participants, Sae-Bae achieved 90% accuracy in identifying who the gestures belonged to.
Her research won third prize in Kaspersky‘s student competition, ‘IT Security for the Next Generation’, earlier this year. This authentication could be less intrusive and more desirable than retina scanning and finger-printing because it uses widely available sensor technologies. Sae-Bae’s next step is to explore the stability and accuracy of individual’s hand gesture over time and try to take this into consideration.