Supervised GAN Watermarking for Intellectual Property Protection
This addresses the need for ownership verification in GAN-generated content, but it is incremental as it builds on existing watermarking and fine-tuning techniques.
The authors tackled the problem of protecting the intellectual property of Generative Adversarial Networks (GANs) by proposing a watermarking method that embeds an invisible signature into generated images, achieving effective embedding and robustness against post-processing like JPEG compression and noise addition.
We propose a watermarking method for protecting the Intellectual Property (IP) of Generative Adversarial Networks (GANs). The aim is to watermark the GAN model so that any image generated by the GAN contains an invisible watermark (signature), whose presence inside the image can be checked at a later stage for ownership verification. To achieve this goal, a pre-trained CNN watermarking decoding block is inserted at the output of the generator. The generator loss is then modified by including a watermark loss term, to ensure that the prescribed watermark can be extracted from the generated images. The watermark is embedded via fine-tuning, with reduced time complexity. Results show that our method can effectively embed an invisible watermark inside the generated images. Moreover, our method is a general one and can work with different GAN architectures, different tasks, and different resolutions of the output image. We also demonstrate the good robustness performance of the embedded watermark against several post-processing, among them, JPEG compression, noise addition, blurring, and color transformations.