How to fool facial recognition systems

Published August 4, 2021

Print

Analytics India magazine quoted Siwei Lyu, Empire Innovation Professor of Computer Science, for a story on attempts to outwit facial recognition software.

Lyu said when a deepfake algorithm generates facial expressions, the new images don’t always map onto the person’s head, or the lighting conditions, or the distance to the camera. Such images have to be geometrically transformed and the process leaves digital footprints allowing researchers to detect the fake videos.

Read the story here.