AS-FIBA: Adaptive Selective Frequency-Injection for Backdoor Attack on Deep Face Restoration

arXiv:2403.06430v1 Announce Type: new
Abstract: Deep learning-based face restoration models, increasingly prevalent in smart devices, have become targets for sophisticated backdoor attacks. These attacks, through subtle trigger injection into input face images, can lead to unexpected restoration outcomes. Unlike conventional methods focused on classification tasks, our approach introduces a unique degradation objective tailored for attacking restoration models. Moreover, we propose the Adaptive Selective Frequency Injection Backdoor Attack (AS-FIBA) framework, employing a neural network for input-specific trigger generation in the frequency …

abstract arxiv attacks backdoor become classification cs.cv deep learning devices face images smart targets tasks through type