Detecting Adversarial Faces Using Only Real Face Self-Perturbations
This addresses security and robustness issues in facial recognition systems by providing a defense that is agnostic to specific attacks and systems, though it is incremental as it builds on existing detection paradigms.
The paper tackles the problem of detecting adversarial faces in facial recognition systems by proposing a method that trains a detector using only real faces and their self-perturbations, achieving generalization to unseen attacks with validation on datasets like LFW and CelebA-HQ against eight gradient-based and two GAN-based attacks.
Adversarial attacks aim to disturb the functionality of a target system by adding specific noise to the input samples, bringing potential threats to security and robustness when applied to facial recognition systems. Although existing defense techniques achieve high accuracy in detecting some specific adversarial faces (adv-faces), new attack methods especially GAN-based attacks with completely different noise patterns circumvent them and reach a higher attack success rate. Even worse, existing techniques require attack data before implementing the defense, making it impractical to defend newly emerging attacks that are unseen to defenders. In this paper, we investigate the intrinsic generality of adv-faces and propose to generate pseudo adv-faces by perturbing real faces with three heuristically designed noise patterns. We are the first to train an adv-face detector using only real faces and their self-perturbations, agnostic to victim facial recognition systems, and agnostic to unseen attacks. By regarding adv-faces as out-of-distribution data, we then naturally introduce a novel cascaded system for adv-face detection, which consists of training data self-perturbations, decision boundary regularization, and a max-pooling-based binary classifier focusing on abnormal local color aberrations. Experiments conducted on LFW and CelebA-HQ datasets with eight gradient-based and two GAN-based attacks validate that our method generalizes to a variety of unseen adversarial attacks.