Saved in:
Bibliographic Details
Main Authors: Shih, Chun-Yen, Peng, Li-Xuan, Liao, Jia-Wei, Chu, Ernie, Chou, Cheng-Fu, Chen, Jun-Cheng
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.11810
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912232642707456
author Shih, Chun-Yen
Peng, Li-Xuan
Liao, Jia-Wei
Chu, Ernie
Chou, Cheng-Fu
Chen, Jun-Cheng
author_facet Shih, Chun-Yen
Peng, Li-Xuan
Liao, Jia-Wei
Chu, Ernie
Chou, Cheng-Fu
Chen, Jun-Cheng
contents Diffusion Models have emerged as powerful generative models for high-quality image synthesis, with many subsequent image editing techniques based on them. However, the ease of text-based image editing introduces significant risks, such as malicious editing for scams or intellectual property infringement. Previous works have attempted to safeguard images from diffusion-based editing by adding imperceptible perturbations. These methods are costly and specifically target prevalent Latent Diffusion Models (LDMs), while Pixel-domain Diffusion Models (PDMs) remain largely unexplored and robust against such attacks. Our work addresses this gap by proposing a novel attack framework, AtkPDM. AtkPDM is mainly composed of a feature representation attacking loss that exploits vulnerabilities in denoising UNets and a latent optimization strategy to enhance the naturalness of adversarial images. Extensive experiments demonstrate the effectiveness of our approach in attacking dominant PDM-based editing methods (e.g., SDEdit) while maintaining reasonable fidelity and robustness against common defense methods. Additionally, our framework is extensible to LDMs, achieving comparable performance to existing approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11810
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Pixel Is Not a Barrier: An Effective Evasion Attack for Pixel-Domain Diffusion Models
Shih, Chun-Yen
Peng, Li-Xuan
Liao, Jia-Wei
Chu, Ernie
Chou, Cheng-Fu
Chen, Jun-Cheng
Computer Vision and Pattern Recognition
Diffusion Models have emerged as powerful generative models for high-quality image synthesis, with many subsequent image editing techniques based on them. However, the ease of text-based image editing introduces significant risks, such as malicious editing for scams or intellectual property infringement. Previous works have attempted to safeguard images from diffusion-based editing by adding imperceptible perturbations. These methods are costly and specifically target prevalent Latent Diffusion Models (LDMs), while Pixel-domain Diffusion Models (PDMs) remain largely unexplored and robust against such attacks. Our work addresses this gap by proposing a novel attack framework, AtkPDM. AtkPDM is mainly composed of a feature representation attacking loss that exploits vulnerabilities in denoising UNets and a latent optimization strategy to enhance the naturalness of adversarial images. Extensive experiments demonstrate the effectiveness of our approach in attacking dominant PDM-based editing methods (e.g., SDEdit) while maintaining reasonable fidelity and robustness against common defense methods. Additionally, our framework is extensible to LDMs, achieving comparable performance to existing approaches.
title Pixel Is Not a Barrier: An Effective Evasion Attack for Pixel-Domain Diffusion Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.11810