INR-Based Generative Steganography by Point Cloud Representation

TL;DR

This work introduces INR-GSPC, an INR-based generative steganography framework that uses a function generator to produce continuous, resolution-independent cover-media represented as point clouds. By sampling point clouds from the generated INR and employing a fixed, point-cloud–aware extractor, the method unifies data formats, eliminates extractor transmission, and supports universal data types beyond fixed grids. The approach delivers exceptional stego-image quality (PSNR > $PSNR>65$) and high message-extraction accuracy (>$Acc>99\%$), while remaining visually indistinguishable and resilient to typical steganalysis and channel distortions. The findings demonstrate that point-cloud representations and INR can significantly enhance the universality, efficiency, and security of generative steganography with multi-resolution capabilities.

Abstract

Generative steganography (GS) directly generates stego-media through secret message-driven generation. It makes the hiding capacity of GS higher than that of traditional steganography, as well as more resistant to classical steganalysis. However, the generators and extractors of existing GS methods can only target specific formats and types of data and lack of universality. Besides, the model size is usually related to the underlying grid resolution, and the transmission behavior of the extractor is susceptible to suspicion of steganalysis. Implicit neural representation(INR) is a technique for representing data in a continuous manner. Inspired by this, we propose an INR-based generative steganography by point cloud representation (INR-GSPC). By using the function generator, the problem of the generator model size growing exponentially with the increase of gridded data has been solved. That is able to generate a wide range of data types and break through the limitation of resolution. In order to unify the data formats of the generator and message extractor, the data is converted to point cloud representation. We designed and fixed a point cloud message extractor. By iterating over the point cloud with adding small perturbations to generate stego-media. This method can avoid the training and transmission process of the message extractor. To the best of our knowledge, this is the first method to apply point cloud to generative steganography. Experiments demonstrate that the stego-images generated by the scheme have an average PSNR value of more than 65, and the accuracy of message extraction reaches more than 99%.

Figures (14)

• Figure 1: The whole framework, showing the scheme process from the sender and receiver.
• Figure 2: Function Generator.
• Figure 3: The Training Process of Function Generator.
• Figure 4: Convolution Neighborhood for Regular Convolutions and PointConv.
• Figure 5: Training Process of Fixed Point Cloud Message Extractor.