Improving the Transferability of Adversarial Attacks on Face Recognition with Beneficial Perturbation Feature Augmentation

TL;DR

This paper introduces BPFA, a novel adversarial attack method that enhances transferability on face recognition models by utilizing beneficial perturbations and hard model concepts, exposing vulnerabilities more effectively.

Contribution

The paper proposes BPFA, a new attack technique that improves adversarial transferability by generating hard models and applying beneficial perturbations during feature map updates.

Findings

BPFA significantly improves attack transferability on face recognition models.

Utilizing hard models reduces overfitting of adversarial examples.

Extensive experiments confirm BPFA's effectiveness in exposing model vulnerabilities.

Abstract

Face recognition (FR) models can be easily fooled by adversarial examples, which are crafted by adding imperceptible perturbations on benign face images. The existence of adversarial face examples poses a great threat to the security of society. In order to build a more sustainable digital nation, in this paper, we improve the transferability of adversarial face examples to expose more blind spots of existing FR models. Though generating hard samples has shown its effectiveness in improving the generalization of models in training tasks, the effectiveness of utilizing this idea to improve the transferability of adversarial face examples remains unexplored. To this end, based on the property of hard samples and the symmetry between training tasks and adversarial attack tasks, we propose the concept of hard models, which have similar effects as hard samples for adversarial attack tasks. Utilizing the concept of hard models, we propose a novel attack method called Beneficial Perturbation Feature Augmentation Attack (BPFA), which reduces the overfitting of adversarial examples to surrogate FR models by constantly generating new hard models to craft the adversarial examples. Specifically, in the backpropagation, BPFA records the gradients on pre-selected feature maps and uses the gradient on the input image to craft the adversarial example. In the next forward propagation, BPFA leverages the recorded gradients to add beneficial perturbations on their corresponding feature maps to increase the loss. Extensive experiments demonstrate that BPFA can significantly boost the transferability of adversarial attacks on FR.