TrustMark: Universal Watermarking for Arbitrary Resolution Images
This work addresses copyright protection and misinformation prevention, particularly for generative AI, with incremental improvements in robustness and flexibility for arbitrary resolutions.
The authors tackled the problem of imperceptible digital watermarking for arbitrary resolution images by proposing TrustMark, a GAN-based method with novel architecture and loss designs to balance image quality and watermark recovery accuracy, achieving state-of-the-art performance on three benchmarks.
Imperceptible digital watermarking is important in copyright protection, misinformation prevention, and responsible generative AI. We propose TrustMark - a GAN-based watermarking method with novel design in architecture and spatio-spectra losses to balance the trade-off between watermarked image quality with the watermark recovery accuracy. Our model is trained with robustness in mind, withstanding various in- and out-place perturbations on the encoded image. Additionally, we introduce TrustMark-RM - a watermark remover method useful for re-watermarking. Our methods achieve state-of-art performance on 3 benchmarks comprising arbitrary resolution images.