Cognitec Adapts Biometric Engine to Masked Faces

Facial recognition company Cognitec has announced that its FaceVACS biometric engine has been updated and is now able to recognize individuals wearing masks.

According to a press release announcing the upgrade, the new capabilities for FaceVACS come via advanced deep learning methods and a new face finder, with the solution showing higher accuracy when confronted with “strong poses and extreme mouth expressions.”

Due to the outbreak of COVID-19, facial recognition systems around the world have faced new authentication challenges as a result of the growth in mask use as efforts to slow the spread of the virus have mandated various safety protocols.

“Cognitec’s algorithm development is responding to our customers’ critical need for accurate identification and verification products in the current pandemic environment,” said Cognitec Managing Director Alfredo Herrera.

Research and development on the part of Cognitec have resulted in an enhancement to the company’s face finder technology, which has in turn yielded detection rates that are up to ten times more accurate in identifying masked faces than previous iterations of the system.

Cognitec also notes that a specialized algorithm has been developed that consumes less memory on Linux-on-Arm and Android platforms, with tests showing a particularly high accuracy for images of people wearing face masks on these platforms.

The company also stated that it is in the process of preparing the release of its next matching algorithm that it expects will further improve accuracy for general matching tasks, with the focus once again being on its ability to identify partially covered faces.

–

November 3, 2020 – by Tony Bitzionis