A reader remarks on the statistical size of the study of how to fool facial recognition systems:
Yeah that’s a pretty small sample. Did the article talk about the makeup? Because I’m pretty sure it’s not the usual “fancy your face up” treatment, it’s something else entirely.
Yes, they did:
What makes the system so clever, says Mariann Hardey at Durham University, UK, is that it doesn’t rely on gaudy colour palettes. Instead, the adversarial system [the AI used to fool the facial recognition system] is limited to using natural make-up hues. That’s important because it lets people try to simultaneously avoid recognition while not drawing attention to themselves: pre-existing research shows wearing outlandishly patterned clothes can foil such systems, but they look obviously like an attempt to avoid detection.
Or, as another reader notes:
Anyone who does theatrical makeup will agree with that. What you see isn’t necessarily what you’ve got …. 🙂
