An Undetectable Watermark for Generative Image Models

TL;DR

This paper introduces the first undetectable watermarking scheme for generative image models that maintains image quality and is robust against removal attacks, enabling secure attribution of generated images.

Contribution

The authors propose a novel undetectable watermarking method for generative images using pseudorandom error-correcting codes, ensuring both undetectability and robustness.

Findings

Watermarks do not degrade image quality.

The scheme is robust against removal attacks.

It can encode up to 512 bits reliably.

Abstract

We present the first undetectable watermarking scheme for generative image models. Undetectability ensures that no efficient adversary can distinguish between watermarked and un-watermarked images, even after making many adaptive queries. In particular, an undetectable watermark does not degrade image quality under any efficiently computable metric. Our scheme works by selecting the initial latents of a diffusion model using a pseudorandom error-correcting code (Christ and Gunn, 2024), a strategy which guarantees undetectability and robustness. We experimentally demonstrate that our watermarks are quality-preserving and robust using Stable Diffusion 2.1. Our experiments verify that, in contrast to every prior scheme we tested, our watermark does not degrade image quality. Our experiments also demonstrate robustness: existing watermark removal attacks fail to remove our watermark from images without significantly degrading the quality of the images. Finally, we find that we can robustly encode 512 bits in our watermark, and up to 2500 bits when the images are not subjected to watermark removal attacks. Our code is available at https://github.com/XuandongZhao/PRC-Watermark.