Robust Message Embedding via Attention Flow-Based Steganography

TL;DR

This paper introduces RMSteg, a novel message embedding framework using transformer-inspired tokenization and normalizing flows, achieving robust, high-quality, and imperceptible message hiding in images resilient to real-world distortions.

Contribution

It presents the first integration of transformer-based tokenization with normalizing flow models for robust image steganography, enabling accurate message recovery under various distortions.

Findings

RMSteg effectively embeds QR codes into images with imperceptible changes.

The method maintains message accuracy even after image printing and photographing.

Experimental results demonstrate superior robustness and image quality compared to existing techniques.

Abstract

Image steganography can hide information in a host image and obtain a stego image that is perceptually indistinguishable from the original one. This technique has tremendous potential in scenarios like copyright protection, information retrospection, etc. Some previous studies have proposed to enhance the robustness of the methods against image disturbances to increase their applicability. However, they generally cannot achieve a satisfying balance between the steganography quality and robustness. Instead of image-in-image steganography, we focus on the issue of message-in-image embedding that is robust to various real-world image distortions. This task aims to embed information into a natural image and the decoding result is required to be completely accurate, which increases the difficulty of data concealing and revealing. Inspired by the recent developments in transformer-based vision models, we discover that the tokenized representation of image is naturally suitable for steganography task. In this paper, we propose a novel message embedding framework, called Robust Message Steganography (RMSteg), which is competent to hide message via QR Code in a host image based on an normalizing flow-based model. The stego image derived by our method has imperceptible changes and the encoded message can be accurately restored even if the image is printed out and photoed. To our best knowledge, this is the first work that integrates the advantages of transformer models into normalizing flow. Our experiment result shows that RMSteg has great potential in robust and high-quality message embedding.