Researchers at Fudan University in China, the Chinese University of Hong Kong, Indiana University, and Alibaba Inc. have created a baseball hat that can fool facial recognition software. Using a technique called “adversarial learning,” the inside of the hat projects infrared dots from LEDs onto strategic locations on the wearer’s face. These dots aren’t visible to the naked eye, but they’ll be picked up by facial recognition systems. This subtly alters how the features in those “strategic spots” appear to facial recognition cameras, causing the system to misidentify the face.
Seventy percent of the time, the new technology was able to trick facial-recognition system FaceNet into identifying hat-wearers as celebrities. The researchers posited that the face-scanning lights could be hidden in an umbrella and possibly even under hair or a wig.
“Based on our findings and attacks, we conclude that face recognition techniques today are still far from secure and reliable when being applied to critical scenarios like authentication and surveillance,” the researchers wrote. “Researchers should pay more attention to the threat from infrared.”
Other recent research has found that using 3D printed glasses can help hide your face well enough to avoid detection from facial recognition software, while another showed that simple stickers on stop signs can trick software into seeing them as speed limit signs
This “adversarial” technology is controversial in that, on one hand, it protects privacy; however, facial recognition is also used to keep citizens safe. Researchers have passed on their findings to the Transportation Security Administration (TSA), and recommended that the TSA consider asking passengers to remove items like glasses and jewelry in the future, since these could be used to beat even state-of-the-art recognition systems..