Sberbank and a global leader in computer vision and machine learning systems VisionLabs (part of the bank’s ecosystem) have studied if face masks can affect security and face recognition, especially in large cities.

The number of people wearing masks has grown due to the coronavirus pandemic, Sberbank notes in a press statement. Can biometric systems still identify the client, if half of their face is hidden?

“Our technology can identify users even if they are wearing face masks. VisionLabs had included that function in the technology back in 2018 and implemented over 20 projects in South East Asia and Japan, where face mask is part of daily routine.

Obviously, the mask complicates the process, but we have learned to identify with high accuracy faces with fake moustache or beard, glasses, scarf, hat, mask or other accessories. Our algorithms take into account a large number of facial features during authentication. Partial cover of the face affects certain algorithms, but the others consistently identify the visible features,” said Alexander Khanin, co-founder and CEO at VisionLabs.

Rambler portal (part of Sberbank’s ecosystem) carried out a survey about Russians’ opinion on face masks on March 5-10. Only 3% of the surveyed said they regularly used masks to prevent infection, 9% said they used masks occasionally or when they were ill, and 88% of the surveyed said they never used masks.

The partner of Fintech section is