Off-The-Shelf Image-to-Image Models Are All You Need To Defeat Image Protection Schemes

2026-02-25

AI Summary

Existing image protection methods rely on imperceptible perturbations to prevent misuse, yet their robustness against general-purpose generative AI attacks remains inadequately validated. This work demonstrates for the first time that off-the-shelf image-to-image generative models—such as Stable Diffusion—combined with simple text prompts can function as universal “denoisers” capable of effectively removing diverse protective perturbations without requiring customized attack strategies. The proposed approach successfully bypasses defenses across six state-of-the-art protection schemes in eight distinct scenarios, outperforming specialized attack methods while preserving image usability for adversaries. These findings reveal a pervasive vulnerability in current image protection mechanisms when confronted with readily available generative AI tools.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Responsible Web: Measurement, analysis, and circumvention of Web censorship

Abstract

Advances in Generative AI (GenAI) have led to the development of various protection strategies to prevent the unauthorized use of images. These methods rely on adding imperceptible protective perturbations to images to thwart misuse such as style mimicry or deepfake manipulations. Although previous attacks on these protections required specialized, purpose-built methods, we demonstrate that this is no longer necessary. We show that off-the-shelf image-to-image GenAI models can be repurposed as generic ``denoisers" using a simple text prompt, effectively removing a wide range of protective perturbations. Across 8 case studies spanning 6 diverse protection schemes, our general-purpose attack not only circumvents these defenses but also outperforms existing specialized attacks while preserving the image's utility for the adversary. Our findings reveal a critical and widespread vulnerability in the current landscape of image protection, indicating that many schemes provide a false sense of security. We stress the urgent need to develop robust defenses and establish that any future protection mechanism must be benchmarked against attacks from off-the-shelf GenAI models. Code is available in this repository: https://github.com/mlsecviswanath/img2imgdenoiser

Problem

• image protection
• generative AI
• adversarial perturbations
• deepfake
• style mimicry

Innovation

• off-the-shelf GenAI
• image protection
• image-to-image translation
• adversarial denoising
• prompt-based attack