SERUM: Simple, Efficient, Robust, and Unifying Marking for Diffusion-based Image Generation

2026-03-11

AI Summary

This work addresses the challenge of copyright protection for images generated by diffusion models by proposing an efficient, robust, and visually lossless watermarking method. The approach embeds a unique watermark signal directly into the initial diffusion noise and employs a lightweight detector for reliable identification. Its key innovation lies in a decoupled architecture that enables personalized watermarking for multiple users with minimal mutual interference. While preserving high generation quality, the method demonstrates significantly enhanced robustness against common image manipulations—including enhancement, compression, and deliberate watermark removal—achieving state-of-the-art true positive rates at a 1% false positive rate. Furthermore, it offers fast embedding and detection speeds alongside low training overhead, making it practical for real-world deployment.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

We propose SERUM: an intriguingly simple yet highly effective method for marking images generated by diffusion models (DMs). We only add a unique watermark noise to the initial diffusion generation noise and train a lightweight detector to identify watermarked images, simplifying and unifying the strengths of prior approaches. SERUM provides robustness against any image augmentations or watermark removal attacks and is extremely efficient, all while maintaining negligible impact on image quality. In contrast to prior approaches, which are often only resilient to limited perturbations and incur significant training, injection, and detection costs, our SERUM achieves remarkable performance, with the highest true positive rate (TPR) at a 1% false positive rate (FPR) in most scenarios, along with fast injection and detection and low detector training overhead. Its decoupled architecture also seamlessly supports multiple users by embedding individualized watermarks with little interference between the marks. Overall, our method provides a practical solution to mark outputs from DMs and to reliably distinguish generated from natural images.

Problem

• watermarking
• diffusion models
• image generation
• robustness
• image authentication

Innovation

• diffusion models
• image watermarking
• robust detection
• efficient marking
• decoupled architecture