A Grid-Level Face Mask Attack is a specific type of adversarial attack targeting face recognition systems. In this attack, subtle perturbations are applied to the pixels at a grid level on an image of a person wearing a face mask. The goal is to cause a face recognition system to misclassify or fail to recognize the person despite the presence of a mask.

The grid-level approach allows attackers to subtly modify pixel values in specific regions of the image, affecting the facial features that the recognition system relies on. Despite the modifications being imperceptible to the human eye, they can lead to misclassification or a failure to recognize the person by the targeted face recognition model.


FACESEC: A Fine-grained Robustness Evaluation Framework for Face Recognition Systems

We present FACESEC, a framework for fine-grained robustness evaluation of face recognition systems. FACESEC evaluation is performed along four dimensions of adversarial modeling: the nature of perturbation (e.g., pixel-level or face accessories), the attacker’s system knowledge (about training data and learning architecture), goals (dodging or impersonation), and capability (tailored to individual inputs or across sets of these). We use FACESEC to study five face recognition systems in both closed-set and open-set settings, and to evaluate the state-of-the-art approach for defending against physically realizable attacks on these. We find that accurate knowledge of neural architecture is significantly more important than knowledge of the training data in black-box attacks. Moreover, we observe that open-set face recognition systems are more vulnerable than closed-set systems under different types of attacks. The efficacy of attacks for other threat model variations, however, appears highly dependent on both the nature of perturbation and the neural network architecture. For example, attacks that involve adversarial face masks are usually more potent, even against adversarially trained models, and the ArcFace architecture tends to be more robust than the others.