Removing Box-Free Watermarks for Image-to-Image Models via Query-Based Reverse Engineering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: An, Haonan, Hua, Guang, Cao, Hangcheng, Fang, Zhengru, Xu, Guowen, Rahardja, Susanto, Fang, Yuguang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909703232028672
author An, Haonan
Hua, Guang
Cao, Hangcheng
Fang, Zhengru
Xu, Guowen
Rahardja, Susanto
Fang, Yuguang
author_facet An, Haonan
Hua, Guang
Cao, Hangcheng
Fang, Zhengru
Xu, Guowen
Rahardja, Susanto
Fang, Yuguang
contents The intellectual property of deep generative networks (GNets) can be protected using a cascaded hiding network (HNet) which embeds watermarks (or marks) into GNet outputs, known as box-free watermarking. Although both GNet and HNet are encapsulated in a black box (called operation network, or ONet), with only the generated and marked outputs from HNet being released to end users and deemed secure, in this paper, we reveal an overlooked vulnerability in such systems. Specifically, we show that the hidden GNet outputs can still be reliably estimated via query-based reverse engineering, leaking the generated and unmarked images, despite the attacker's limited knowledge of the system. Our first attempt is to reverse-engineer an inverse model for HNet under the stringent black-box condition, for which we propose to exploit the query process with specially curated input images. While effective, this method yields unsatisfactory image quality. To improve this, we subsequently propose an alternative method leveraging the equivalent additive property of box-free model watermarking and reverse-engineering a forward surrogate model of HNet, with better image quality preservation. Extensive experimental results on image processing and image generation tasks demonstrate that both attacks achieve impressive watermark removal success rates (100%) while also maintaining excellent image quality (reaching the highest PSNR of 34.69 dB), substantially outperforming existing attacks, highlighting the urgent need for robust defensive strategies to mitigate the identified vulnerability in box-free model watermarking.
format Preprint
id arxiv_https___arxiv_org_abs_2507_18034
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Removing Box-Free Watermarks for Image-to-Image Models via Query-Based Reverse Engineering
An, Haonan
Hua, Guang
Cao, Hangcheng
Fang, Zhengru
Xu, Guowen
Rahardja, Susanto
Fang, Yuguang
Cryptography and Security
The intellectual property of deep generative networks (GNets) can be protected using a cascaded hiding network (HNet) which embeds watermarks (or marks) into GNet outputs, known as box-free watermarking. Although both GNet and HNet are encapsulated in a black box (called operation network, or ONet), with only the generated and marked outputs from HNet being released to end users and deemed secure, in this paper, we reveal an overlooked vulnerability in such systems. Specifically, we show that the hidden GNet outputs can still be reliably estimated via query-based reverse engineering, leaking the generated and unmarked images, despite the attacker's limited knowledge of the system. Our first attempt is to reverse-engineer an inverse model for HNet under the stringent black-box condition, for which we propose to exploit the query process with specially curated input images. While effective, this method yields unsatisfactory image quality. To improve this, we subsequently propose an alternative method leveraging the equivalent additive property of box-free model watermarking and reverse-engineering a forward surrogate model of HNet, with better image quality preservation. Extensive experimental results on image processing and image generation tasks demonstrate that both attacks achieve impressive watermark removal success rates (100%) while also maintaining excellent image quality (reaching the highest PSNR of 34.69 dB), substantially outperforming existing attacks, highlighting the urgent need for robust defensive strategies to mitigate the identified vulnerability in box-free model watermarking.
title Removing Box-Free Watermarks for Image-to-Image Models via Query-Based Reverse Engineering
topic Cryptography and Security
url https://arxiv.org/abs/2507.18034