Social Vision Group

Real-world application

Some of our research has an applied focus, including work on face masks, which we started during the COVID-19 pandemic, and our work on face recognition algorithms.

Algorithms and humans perform differently when given images of real vs. superimposed masks.

From our research: Ritchie et al. (2024)

Facial recognition technology used by police is now very accurate – but public understanding lags behindIt’s a common misconception that facial recognition technology captures and stores an image of your face.

Why we can still recognise people in face masksEven though they cover most of our features, face masks are no match for our highly-evolved capacity to recognise friends.

Voter ID: most people are terrible at matching faces to photos, making polling checks unreliableResearch suggests that photo ID checks at polling stations risk voters being turned away because of errors.

Google Sites

Report abuse