Current user-identification techniques
are often based on logins and PIN
codes or passwords. The overhead of
logging in was more acceptable when
we used to sit down for long sessions
of work in front of a PC, but it now
has become a nuisance—to the point
where many users forgo any authentication
on their mobile devices. Biometric
techniques, such as fingerprint readers
or face recognition, are more resource
intensive and face their own usability
challenges. Fingerprint readers require
space on a device, add cost, and can be
difficult to use in low-humidity environments.
Face recognition has higher
processing requirements and can be difficult
in low-lighting conditions.
Our goal has therefore been to find
a less obtrusive way to identify users
during their regular interactions with