A Brief Survey on Deep Learning Based Data Hiding

TL;DR

This paper provides a comprehensive review of deep learning-based data hiding techniques, classifying them by capacity, security, and robustness, and discusses architectures, applications, and adversarial attack perspectives.

Contribution

It offers a structured classification and summary of deep hiding methods, including architectures and application strategies, with insights into adversarial attacks.

Findings

Classified deep hiding methods by capacity, security, robustness

Summarized architectures and application strategies

Discussed adversarial attack implications

Abstract

Data hiding is the art of concealing messages with limited perceptual changes. Recently, deep learning has enriched it from various perspectives with significant progress. In this work, we conduct a brief yet comprehensive review of existing literature for deep learning based data hiding (deep hiding) by first classifying it according to three essential properties (i.e., capacity, security and robustness), and outline three commonly used architectures. Based on this, we summarize specific strategies for different applications of data hiding, including basic hiding, steganography, watermarking and light field messaging. Finally, further insight into deep hiding is provided by incorporating the perspective of adversarial attack.