Vanishing Watermarks: Diffusion-Based Image Editing Undermines Robust Invisible Watermarking

TL;DR

This paper reveals that diffusion-based image editing techniques can effectively erase robust invisible watermarks, exposing a fundamental vulnerability and highlighting the need for new resilient watermarking strategies.

Contribution

It provides a comprehensive theoretical and empirical analysis showing how diffusion models can remove watermarks and introduces a guided diffusion attack targeting watermark signals.

Findings

Diffusion models can erase watermarks while preserving image quality.

Watermark detectability drops to near zero after diffusion-based edits.

Current watermarking methods are vulnerable to generative model-based attacks.

Abstract

Robust invisible watermarking schemes aim to embed hidden information into images such that the watermark survives common manipulations. However, powerful diffusion-based image generation and editing techniques now pose a new threat to these watermarks. In this paper, we present a comprehensive theoretical and empirical analysis demonstrating that diffusion models can effectively erase robust watermarks even when those watermarks were designed to withstand conventional distortions. We show that a diffusion-driven image regeneration process, which leverages generative models to recreate an image, can remove embedded watermarks while preserving the image's perceptual content. Furthermore, we introduce a guided diffusion-based attack that explicitly targets the embedded watermark signal during generation, significantly degrading watermark detectability. Theoretically, we prove that as an image undergoes sufficient diffusion transformations, the mutual information between the watermarked image and the hidden payload approaches zero, leading to inevitable decoding failure. Experimentally, we evaluate multiple state-of-the-art watermarking methods (including deep learning-based schemes like StegaStamp, TrustMark, and VINE) and demonstrate that diffusion edits yield near-zero watermark recovery rates after attack, while maintaining high visual fidelity of the regenerated images. Our findings reveal a fundamental vulnerability in current robust watermarking techniques against generative model-based edits, underscoring the need for new strategies to ensure watermark resilience in the era of powerful diffusion models.