The neural network was taught to create “universal” faces to deceive identification systems

Israeli scientists have developed the StyleGAN Generative Adversarial Network (GAN), which is capable of generating “master faces” (by analogy with master keys). Each of the images can simulate a large number of personalities for recognition systems.

As the authors of the study assure, 9 synthesized faces are able to replace images of at least 40% of people from an open database. During the experiment, scientists tested the StyleGAN Generative Adversarial Network (GAN) neural network on three effective face recognition systems. The research was carried out jointly with scientific institutions in Tel Aviv.

In the course of the work, the scientists found out that a single generated face is capable of imitating 20% ​​of faces from the open database of the University of Massachusetts. As you know, it is she who is often used to test personality recognition systems.

The method of Israeli scientists makes it possible to use open sources as “models” to “replace” the overwhelming majority of people without using closed databases. Under different conditions, scientists were able to achieve “positive” identification of more than 40% to 60% of faces using only 9 generated photographs.

The system uses the so-called. An “evolutionary algorithm” and a “neuropredictor” that estimates the likelihood of how much the current “candidate” will be better than the faces generated during previous attempts.

If you have found a spelling error, please, notify us by selecting that text and pressing Ctrl+Enter.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Alexandr Ivanov earned his Licentiate Engineer in Systems and Computer Engineering from the Free International University of Moldova. Since 2013, Alexandr has been working as a freelance web programmer.
Function: Web Developer and Editor
Alexandr Ivanov

Spelling error report

The following text will be sent to our editors: