Company prints 3D mask that fools facial recognition systems

Date:17 December 2019 Author: Kyro Mitchell Tags:, , ,

As technology has improved over the years, we’ve become more reliant on automated security systems such as figure print scanners and facial recognition to help us go about our day-to-day lives. Seeing as these technologies are now being used in places like airports and government buildings, the question of how easy it is to fool a facial recognition system has been brought up by Kneron, a leading facial recognition company from San Diego.

To answer this question Kneron printed their very own high-quality 3D mask that accurately resembled another persons face. They then tested the mask across three different platforms in a variety of different countries, and to the surprise of everyone involved the mask managed to fool the systems.

Researchers from Kneron first used their 3D mask to fool the facial recognition systems on applications like WeChat and AlliPay, granting them the ability to make purchases from another persons account.

They then moved on to the facial recognition systems in China’s train stations where commuters use facial recognition to pay for their journey, and they fooled that system as well.

Finally, the team of researchers moved on to Schiphol, Netherlands largest airport, and this is where things get really scary. They managed to trick Schiphol’s self-boarding system with nothing more than a photo of someone on a phone screen, according to Fortune.

“The technology is available to fix these issues but firms have not upgraded it. They are taking shortcuts at the expense of security.” CEO of Kneron, Albert Liu told Fortune.

All is not lost for facial recognition systems though, as Kneron announced they were unable to fool the iPhone and Huawei’s advanced facial recognition scanners. However, the fact that they we able to fool systems in airports and railways stations points to the fact that 100% secure facial ID systems are a long way off.

Feature image: Twitter/Chris Gledhill

Latest Issue :

May-June 2022