Using a GAN to Generate Adversarial Examples to Facial Image Recognition
This work addresses privacy concerns for individuals whose images are posted online, but it is incremental as it builds on existing adversarial example techniques.
The authors tackled the problem of protecting facial images from being used by unauthorized facial recognition systems by generating adversarial examples using a GAN, achieving an acceptable success rate in fooling recognition and reducing training time by removing the discriminator component.
Images posted online present a privacy concern in that they may be used as reference examples for a facial recognition system. Such abuse of images is in violation of privacy rights but is difficult to counter. It is well established that adversarial example images can be created for recognition systems which are based on deep neural networks. These adversarial examples can be used to disrupt the utility of the images as reference examples or training data. In this work we use a Generative Adversarial Network (GAN) to create adversarial examples to deceive facial recognition and we achieve an acceptable success rate in fooling the face recognition. Our results reduce the training time for the GAN by removing the discriminator component. Furthermore, our results show knowledge distillation can be employed to drastically reduce the size of the resulting model without impacting performance indicating that our contribution could run comfortably on a smartphone