Facial recognition is one of those technologies that was seen as almost science fiction until recently. But while initial tests were inconsistent, to say the least, and there has been pretty substantial pushback from certain circles regarding the moral legitimacy of its use, it’s here to stay and it couldn’t have arrived at a more convenient time.
Since March 2020 and the onset of the coronavirus pandemic, our world has changed in a pretty fundamental way to the extent that contact is now seen as taboo. Freestanding touchscreens have been seen as the safe standard for years now but that was before the world of contact tracing and confusion.
Today, the idea of using a touchscreen might understandably put people off as they worry about touching a surface used by hundreds or thousands of potential virus carriers and the only legitimate alternatives are voice and face recognition, the latter of which we’ll be breaking down below.
What is facial recognition and how does it work?
Have you got the latest iPhone? If the answer is yes then you’ve probably already used facial recognition as it uses the front-facing camera to scan your face for recognisable traits and patterns and unlocks your phone accordingly.
This is, in a nutshell, how facial recognition works – it uses a camera to capture, analyse and compare minute facial details and is considered to be the most natural, reliable and least intrusive means of biometric measurement available today.
Identify and verify
Whilst facial recognition can be used for other purposes its primary function is to identify and authenticate a person wanting access to something. A 2D or 3D camera will first capture a face and turn it into digital data that it then compares with the facial data stored in its database. It can take just a matter of seconds to scan everything from the spacing between your eyes to the contours of your features and transform it into usable data.
It’s also a technology that has come on leaps and bounds in recent years in terms of accuracy and stability, to the extent that it’s now possible to identify a moving face even in a crowded room with the right software and hardware. Indeed, between 2013 and 2018 a report by NIST found that there had been a 20x improvement in the technology due to advances in AI and deep learning that mean facial detection is now faster, more accurate and less expensive.
Why is facial recognition helpful against the spread of COVID-19?
The answer to that is simple enough – it requires no physical interaction from the user. There has, therefore, already been talks happening between various companies and the UK Government regarding using facial recognition technology in connection with so-called “immunity passports” that would effectively be able to scan a person’s face and let the authorities know immediately whether they were COVID positive. The NHS facial-recognition system, built by iProov, requires users to submit a photo of themselves from an official document such as their passport or driving license and it builds an official profile of their face that way.
However, when used at this scale, there is a fair amount of scepticism by those that view this as highly invasive and as something that infringes on personal rights to privacy.
On a more basic level, it’s most powerful potential implication is in retail and office environments. Here, it could be integrated into a solution with a hand sanitiser station, an LED display screen and a high-quality camera to scan customers and match their faces with a service such as the NHS track and trace app and use that information to either allow them access or tell them to go home and self-isolate.
Of course, this is all largely speculative at the moment, but if we are to stand any hope of moving forward as a society and living with this virus, we need to act fast and develop the infrastructure that will allow the new normal to become the normal normal.
Image: metamorworks / Shutterstock.com
About author
