Advancements in Facial Recognition

Technology is constantly evolving and, as we become familiar with new advancements, we begin to rely upon them. 

Many of us own a device that can be accessed via fingerprint ID, and now many phones and tablets also carry facial recognition technology. However, since the pandemic began and face masks became mandatory, users have complained about how difficult it has become to access a device when half of their face is routinely covered.

It’s clearly an issue. According to a NIST study, ‘of the best 89 commercial facial recognition algorithms tested, there were error rates of between 5% and 50% in matching digitally-applied face masks with photos of the same person without a mask.’

I personally rely upon facial recognition to access my phone, my banking app, and more; to me, it’s convenient, quick and easy. However, when I’m out and about in public wearing my mask, I either have to enter a password to unlock my phone or remove my mask so that my face can be scanned. Though it was natural a few years ago to put my password into my phone to unlock it, it seems a chore now. 

Facial recognition on an iPhone only works if the user is looking directly at the device…so, why doesn’t facial recognition work when half our face is covered if it’s our eyes that are the key to unlocking the phone?

Whether it’s a valid hack, users have said that there’s a way to get round the problem on an iPhone. According to an article by www.bbc.co.uk/, research published last year shows that ‘half a face is enough for a specifically-designed facial recognition algorithm to work’. This may mean that sometime soon, our faces could be fully ‘detectable’ even when we wear a face mask. In the meantime, and following this half-face ethos, iPhone users can try folding a mask vertically and hanging both ear loops over one ear (holding the mask so that it covers half your face) then accessing the ‘reset face ID’ setting or ‘alternate appearance’ and scanning this ‘mask half on, mask half off’ version of yourself.

Covid-19 has created so many challenges during 2020, not just in the world of facial recognition. It has also created many opportunities. 

What kind of advancements do you think we could see in the future? Did the iPhone hack work for you? Let us know at @intheknowemag on Twitter.