Blind Deep-Learning-Based Image Watermarking Robust Against Geometric Transformations

arXiv:2402.09062v1 Announce Type: cross
Abstract: Digital watermarking enables protection against copyright infringement of images. Although existing methods embed watermarks imperceptibly and demonstrate robustness against attacks, they typically lack resilience against geometric transformations. Therefore, this paper proposes a new watermarking method that is robust against geometric attacks. The proposed method is based on the existing HiDDeN architecture that uses deep learning for watermark encoding and decoding. We add new noise layers to this architecture, namely for a differentiable JPEG estimation, rotation, …

arxiv attacks copyright cs.cr cs.cv cs.mm digital image images protection resilience robustness watermarking watermarks