RAW: A Robust and Agile Plug-and-Play Watermark Framework for AI-Generated Images with Provable Guarantees
This addresses the need for reliable watermarking in AI-generated images to safeguard intellectual property, though it appears incremental as it builds on existing watermarking methods with improvements in robustness and agility.
The paper tackles the problem of protecting intellectual property and preventing misuse of AI-generated images by introducing a robust and agile plug-and-play watermark detection framework called RAW, which shows a notable increase in AUROC from 0.48 to 0.82 compared to state-of-the-art approaches under adversarial attacks while maintaining image quality.
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.