跳到正文
arXiv cs.AI· Wenbo Jiang, Hongwei Li, Jiaming He, Rui Zhang, Guowen Xu, Tianwei Zhang, Rongxing Lu·· 10 小时前AI 评分31

I2I Backdoor:针对图像到图像任务的后门攻击

I2I Backdoor: Backdoor Attacks against Image-to-Image Tasks

AI 导读

研究者首次系统性地揭示了图像到图像(I2I)网络的后门攻击漏洞,提出 I2I Backdoor 攻击方法:被植入后门的 I2I 网络在干净输入上表现正常,一旦输入图像出现后门触发器,便输出特定的不当图像。

正文 · 原文

View PDF HTML (experimental)

Abstract:With the rapid development of deep learning technology, deep learning-based Image-to-Image (I2I) networks have become the predominant choice for I2I tasks like image super-resolution and denoising. Despite their remarkable performance, the security of I2I networks has not been thoroughly investigated. While some studies have probed their susceptibility to adversarial attacks, none have explored the backdoor attack against I2I networks, which is a more stealthy and severe threat. In this work, for the first time, we comprehensively investigate the vulnerability of I2I networks to backdoor attacks. We propose a backdoor attack against I2I tasks, where the backdoored I2I network behaves normally on clean input images, yet outputs a specific inappropriate image when the backdoor trigger appears on the input image. To achieve such an I2I backdoor attack, we design a universal adversarial perturbation (UAP) generation algorithm for I2I networks, where the generated UAP is used as the trigger for the I2I backdoor. Besides, multi-task learning (MTL) with dynamic weighting methods is employed in the backdoor training process to gain better results. Expanding our focus beyond I2I tasks, we extend our I2I backdoor to attack downstream tasks, including image classification and object detection. Specifically, the backdoor-triggered image processed by the backdoored image denoising network can fool the downstream image classifiers and object detectors. Extensive experiments demonstrate the effectiveness of the I2I backdoor on state-of-the-art I2I network architectures as well as the robustness against different backdoor defenses.
Comments: Accepted for publication in IEEE Transactions on Dependable and Secure Computing (TDSC)
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR)
Cite as: arXiv:2407.10445 [cs.CV]
  (or arXiv:2407.10445v2 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2407.10445

arXiv-issued DOI via DataCite

Journal reference: IEEE Trans. Dependable Secure Comput. 22(6) (2025) 7996-8010
Related DOI: https://doi.org/10.1109/TDSC.2025.3603639

DOI(s) linking to related resources

Submission history

From: Wenbo Jiang [view email]
[v1] Mon, 15 Jul 2024 05:14:17 UTC (3,898 KB)
[v2] Wed, 18 Nov 2026 05:20:45 UTC (3,673 KB)

来源:arXiv cs.AI · arxiv.org