Data Hiding with Deep Learning: A Survey Unifying Digital Watermarking and Steganography

Data hiding is the process of embedding information into a noise-tolerant signal such as a piece of audio, video, or an image, including Digital Watermarking for robust identity verification and Steganography to embed data for the purpose of secure and secret communication. This survey provides a summary of recent advancements in deep learning techniques for data hiding in watermarking and steganography, and categorizes them based on model architectures and noise injection methods. The objective functions, evaluation metrics, and datasets used for training these data hiding models are comprehensively summarised. Finally, we propose and discuss possible future directions for the unification of digital watermarking and steganography in software engineering to promote Responsible AI.