RAW: A Robust and Agile Plug-and-Play Watermark Framework for AI-Generated Images with Provable Guarantees

TL;DR

This paper presents RAW, a flexible watermark detection framework for AI-generated images that uses learnable watermarks and provable guarantees to improve robustness against adversarial attacks.

Contribution

Introduces RAW, a novel plug-and-play watermarking framework with learnable watermarks and provable false positive guarantees, compatible with various generative models.

Findings

Significant increase in AUROC from 0.48 to 0.82 under adversarial attacks.

Maintains image quality with comparable FID and CLIP scores.

Supports on-the-fly watermark injection after training.

Abstract

Safeguarding intellectual property and preventing potential misuse of AI-generated images are of paramount importance. This paper introduces a robust and agile plug-and-play watermark detection framework, dubbed as RAW. As a departure from traditional encoder-decoder methods, which incorporate fixed binary codes as watermarks within latent representations, our approach introduces learnable watermarks directly into the original image data. Subsequently, we employ a classifier that is jointly trained with the watermark to detect the presence of the watermark. The proposed framework is compatible with various generative architectures and supports on-the-fly watermark injection after training. By incorporating state-of-the-art smoothing techniques, we show that the framework provides provable guarantees regarding the false positive rate for misclassifying a watermarked image, even in the presence of certain adversarial attacks targeting watermark removal. Experiments on a diverse range of images generated by state-of-the-art diffusion models reveal substantial performance enhancements compared to existing approaches. For instance, our method demonstrates a notable increase in AUROC, from 0.48 to 0.82, when compared to state-of-the-art approaches in detecting watermarked images under adversarial attacks, while maintaining image quality, as indicated by closely aligned FID and CLIP scores.