from Schneier on Security
The Chinese facial recognition company Hanwang claims it can recognize people wearing masks:
The company now says its masked facial recognition program has reached 95 percent accuracy in lab tests, and even claims that it is more accurate in real life, where its cameras take multiple photos of a person if the first attempt to identify them fails.
[…]
Counter-intuitively, training facial recognition algorithms to recognize masked faces involves throwing data away. A team at the University of Bradford published a study last year showing they could train a facial recognition program to accurately recognize half-faces by deleting parts of the photos they used to train the software.
When a facial recognition program tries to recognize a person, it takes a photo of the person to be identified, and reduces it down to a bundle, or vector, of numbers that describes the relative positions of features on the face.
[…]
Hanwang’s system works for masked faces by trying to guess what all the faces in its existing database of photographs would look like if they were masked.
More here.
The article “Facial Recognition for People Wearing Masks” is incredibly relevant to today’s world. At the beginning of the pandemic, I had an iPhone 7 meaning that I didn’t have to worry about facial recognition being needed to quickly unlock my phone. It wasn’t until last year when I upgraded to the iPhone XR when I found out the problems faced with having facial recognition in a time where masks are mandated. When you have nothing obstructing your face, being able to scan your face to log in to your phone is perfect. Its quick, hands free, and relatively secure. No one else has my face and the technology is advanced enough to make sure that I’m the only person that is able to unlock my phone. Opposed to the fingerprint option on my old iPhone 7, in order to access a lot of my password facial recognition is needed. This means that while in public with a mask on I have to move my mask to be able to scan my face. With that said however, my whole face does not need to be shown in order to trigger the scan. This leads into the possibility of a technology that can identify a person with a mask on. A Chinese company named “Hanwang” claims that their newest facial recognition software has a 95% success rate when it comes to scanning somebody with a mask on. This software works by trying to guess what the person’s face looks like without a mask. Basically, filling in the blanks. In my opinion, I don’t think this is a good idea. I don’t see how they can make it so that it only works with one person. People tend to look alike when masked. This means that unless the software is completely sound, people might be able to get away with unlocking somebody else’s phone. With that said however, I like the idea. If there is a way that this can successfully be done power to them. It just doesn’t seem safe enough to fully trust, especially with something as important as bank info or passwords, things often associated with facial recognition.