Disentangle and Drop (DnD) is proposed, an attack that is agnostic to the watermark method and treats watermark removal as a representation routing problem and shows that one shared operating setting gives competitive removal with high visual fidelity.
Abstract
Invisible image watermarks are commonly evaluated against benign postprocessing operations such as compression, resizing, blur, and color changes. These tests leave out a different threat: a learned remover that preserves semantic image content while discarding residual evidence that carries the payload. We propose Disentangle and Drop (DnD), an attack that is agnostic to the watermark method and treats watermark removal as a representation routing problem. DnD decomposes a watermarked image into a semantic grayscale carrier and an auxiliary residual branch, and then suppresses the residual branch to reduce watermark evidence. The model is trained with latent spectral perturbations and low-strength diffusion exposure so that the drop operation remains stable under adaptive reconstruction. Experiments on seven representative watermark families show that one shared operating setting gives competitive removal with high visual fidelity. Operating scans and ablations separate usable attacks from image-damaging settings: stronger noise or diffusion can raise removal scores by damaging the image, while the practical regime comes from dropping the residual latent. These results argue for evaluating watermark robustness against learned removal at the representation level, not only against conventional image edits.
Invisible image watermark removal has received growing attention. Despite substantial progress, existing attacks face a tension between practicality and specificity. Attacks exploiting detector outputs, decoder responses, or paired images can be tailored to the watermark decision boundary, but require information rarel...
Jun Yao, Chao Wang, Yu-Peng Qiu et al.· 0 citations
Digital watermarking supports source attribution for AI-generated images, but its reliability depends on resistance to removal attacks. Some attacks attempt to remove watermarks by forcing the decoded watermark to differ from the original. However, this can produce an inverted watermark that remains detectable, causing...
Sai-Bo Ye, Hua-Jie Chen, Xin Guo et al.· 0 citations
This work introduces RemTraceNet, which fuses constrained residuals, local relations, FFT/Haar statistics, and block-DCT evidence at native resolution and outperforms retrained SRNet, ZhuNet, and SiaStegNet baselines by 10.80--15.68 percentage points.
This work identifies two complementary laundering regimes: OpenAI models produce the strongest payload disruption across the evaluated schemes, whereas Nano Banana 2 shows that DwtDct remains vulnerable under high-fidelity reconstruction.
This work identifies two mechanisms that strengthen the dependence of watermark evidence on the cover image, thereby suppressing the residual transferability, and introduces CoverLock, a plug-and-play strategy for existing watermarking systems that strengthens such image dependence without architectural redesign.
The proliferation of artificial intelligence-generated content (AIGC) has greatly boosted image synthesis efficiency, yet it also raises serious concerns over copyright infringement and content forgery. Existing watermarking methods generally show weak resistance to diffusion-based regeneration attacks, and struggle to...
Zhi-Tao Han, Siau-Chuin Liew, Anis Farihan Mat Raffei et al.· 2026 IEEE 1st International...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.