Silicon Valley-based UnifyID has announced a new gait biometrics solution that leverages an end user’s mobile device to authenticate them as they walk.
Dubbed ‘GaitAuth’, the solution is available as a lightweight SDK for iOS and Android. It’s designed to passively identify an end user as they walk, with UnifyID claiming that it can do so after only a few steps.
“We’ve tested our models using anonymized data from millions of mobile phones and found the accuracy of gait-based authentication can rival other biometrics like fingerprint, iris, or face,” the company said, adding that GaitAuth is the product of almost four years of R&D.
UnifyID first came out of “stealth mode” in September of 2016. Emerging from the Stanford StartX incubator, the startup initially focused on building a behavioral biometrics platform based on how a user interacts with a mobile device, such as typing speed and touchscreen movements, in addition to metrics tied to gait.
The company has evidently turned all of its attention to gait biometrics in the years since. With UnifyID boasting that its gait biometrics technology can be as accurate as more traditional authentication modalities like facial and fingerprint recognition, the company says GaitAuth can be implemented in a range of use cases, from contactless access control to ATMs to boarding at the airport.
The end user doesn’t need to be walking for a given authentication prompt, either: having authenticated the user as they walked, GaitAuth can tell if the same user is still in possession of the registered mobile device even if they have been standing still or sitting. Likewise, it can tell if the device has been handed off to someone else, UnifyID says.
UnifyID says that GaitAuth is just the first step in its plan to launch multiple APIs enabling “continuous, implicit authentication”, though any further potential product launches are currently under wraps.
(Originally posted on FindBiometrics)