Yet another case of misleading information being spread by tech media
"How Making a Funny Face Could Boost the Security of Face Unlock Features" https://gizmodo.com/how-making-a-funny-face-could-boost-the-security-of-fac-1846502530
Low false acceptance rate is one thing and good security is another.
The information those tech media always fail to report is
1. what the empirical (actually-measured) false rejection rate is corresponding to the empirical false acceptance rate
2. what fallback means is used in case of false rejection
Logic tells that it is absolutely impossible for anybody to judge what level of security ａparticular biometrics solution provides without the answers to these two questions as examined here – “Negative Security Effect of Biometrics Deployed in Cyberspace”
The effect of anti-spoofing measure such as registering funny faces is like a mirage as discussed here – “Spoofing and Liveness-Detection of Biometrics”
How long do we have to live with those misguided and misguiding tech media?
< Videos on YouTube>