StyleMark: A Robust Watermarking Method for Art Style Images Against Black-Box Arbitrary Style Transfer
This addresses copyright infringement for artists by providing a practical solution for tracking style images in AST applications, though it is incremental as it builds on existing watermarking techniques.
The authors tackled the problem of unauthorized use of art style images in arbitrary style transfer (AST) by proposing StyleMark, a robust watermarking method that embeds copyright watermarks into style images to enable precise attribution after AST, achieving significant robustness against black-box AST and common distortions.
Arbitrary Style Transfer (AST) achieves the rendering of real natural images into the painting styles of arbitrary art style images, promoting art communication. However, misuse of unauthorized art style images for AST may infringe on artists' copyrights. One countermeasure is robust watermarking, which tracks image propagation by embedding copyright watermarks into carriers. Unfortunately, AST-generated images lose the structural and semantic information of the original style image, hindering end-to-end robust tracking by watermarks. To fill this gap, we propose StyleMark, the first robust watermarking method for black-box AST, which can be seamlessly applied to art style images achieving precise attribution of artistic styles after AST. Specifically, we propose a new style watermark network that adjusts the mean activations of style features through multi-scale watermark embedding, thereby planting watermark traces into the shared style feature space of style images. Furthermore, we design a distribution squeeze loss, which constrain content statistical feature distortion, forcing the reconstruction network to focus on integrating style features with watermarks, thus optimizing the intrinsic watermark distribution. Finally, based on solid end-to-end training, StyleMark mitigates the optimization conflict between robustness and watermark invisibility through decoder fine-tuning under random noise. Experimental results demonstrate that StyleMark exhibits significant robustness against black-box AST and common pixel-level distortions, while also securely defending against malicious adaptive attacks.