Unified Steganography via Implicit Neural Representation

2025-05-03

AI Summary

Existing steganographic methods are format-specific, severely limiting their generality. To address this, we propose the first cross-modal unified steganography framework based on Implicit Neural Representations (INRs), which jointly encodes secret data and multimodal carriers—such as images, videos, audio, signed distance fields (SDFs), and neural radiance fields (NeRFs)—into shared neurons of an INR network, thereby eliminating format constraints. We introduce a novel steganographic paradigm wherein INR neurons serve as universal information carriers and design a private-key-driven keyed neuron localization strategy to enable key-controllable, format-agnostic embedding and extraction. Through joint multimodal optimization during training, our method achieves high embedding capacity and perceptual imperceptibility while significantly enhancing robustness against steganalysis detection. This work establishes a new paradigm for cross-domain data privacy protection grounded in implicit neural representations.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Security and Privacy: Data transparency and provenance

Abstract

Digital steganography is the practice of concealing for encrypted data transmission. Typically, steganography methods embed secret data into cover data to create stega data that incorporates hidden secret data. However, steganography techniques often require designing specific frameworks for each data type, which restricts their generalizability. In this paper, we present U-INR, a novel method for steganography via Implicit Neural Representation (INR). Rather than using the specific framework for each data format, we directly use the neurons of the INR network to represent the secret data and cover data across different data types. To achieve this idea, a private key is shared between the data sender and receivers. Such a private key can be used to determine the position of secret data in INR networks. To effectively leverage this key, we further introduce a key-based selection strategy that can be used to determine the position within the INRs for data storage. Comprehensive experiments across multiple data types, including images, videos, audio, and SDF and NeRF, demonstrate the generalizability and effectiveness of U-INR, emphasizing its potential for improving data security and privacy in various applications.

Problem

• General framework for multi-type data steganography
• Eliminates need for type-specific steganography designs
• Enhances data security via neural representation and keys

Innovation

• Uses Implicit Neural Representation for steganography
• Employs private key for secret data positioning
• Key-based selection strategy for data storage