A GAN-Based Coverless Video Steganography Utilizing Inter-Frame Similarity

Abstract

GAN-based coverless video steganography method that enhances data security without modifying the carrier media. A Generative Adversarial Network (GAN) is employed to create synthetic video sequences, which are mapped to secret data, forming a Secret Communication Video Database (SCVD). The proposed method generated  inter-frame using GAN and use its  similarity for improved embedding capacity and security. Unlike conventional methods, this approach eliminates the need for auxiliary data transmission, reducing the risk of steganalysis detection. Experimental results demonstrate the superiority of this technique over existing methods in terms of robustness, embedding capacity, and security.