Generative Implicit Steganography via Message Mapping
Source abstract
Abstract
Generative steganography (GS) generates stego-media via secret messages, but existing GS only targets single-type multimedia data with poor universality. The generator and extractor sizes are highly coupled with resolution. Message mapping converts secret messages and noise, yet current GS schemes based on it use gridded data, failing to generate diverse multimedia universally. Inspired by implicit neural representation (INR), we propose generative implicit steganography via message mapping (GIS). We designed single-bit and multi-bit message mapping schemes in function domains. The scheme’s function generator eliminates the coupling between model and gridded data sizes, enabling diverse multimedia generation and breaking resolution limits. A dedicated point cloud extractor is trained for adaptability. Through a literature review, this scheme is the first to perform message mapping in the functional domain. During the experiment, taking images as an example, methods such as PSNR, StegExpose, and neural pruning were used to demonstrate that the generated image quality is almost indistinguishable from the real image. At the same time, the generated image is robust. The accuracy of message extraction can reach 96.88% when the embedding capacity is 1 bpp, 89.84% when the embedding capacity is 2 bpp, and 82.21% when the pruning rate is 0.3.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Steganography and Watermarking Techniques · Digital Media Forensic Detection · Chaos-based Image/Signal Encryption
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Fengdong New City Campus, Engineering University of Chinese People’s Armed Police Force, Xixian New Area, Xi’an 710018, China
References (48)
Show source-provided references
1. Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A.C., and Bengio, Y. (2021, January 6–8). Generative adversarial networks. Proceedings of the 14th International Conference on Computing Communication and Networking Technologies (ICCCNT), Kharagpur, India.
2. Hayes, J., and Danezis, G. (2017, January 4–9). Generating steganographic images via adversarial training. Proceedings of the Neural Information Processing Systems, Long Beach, CA, USA.
3. Kim, D., Shin, C., Choi, J., Jung, D., and Yoon, S. (2023). Diffusion-Stego: Training-free Diffusion Generative Steganography via Message Projection. arXiv.
4. Ho, J., Jain, A., and Abbeel, P. (2020). Denoising diffusion probabilistic models. arXiv.
5. Wei, P., Zhou, Q., Wang, Z., Qian, Z., Zhang, X., and Li, S. (2023). Generative Steganography Diffusion. arXiv.
6. Kingma, D.P., and Dhariwal, P. (2018). Glow: Generative Flow with Invertible 1 × 1 Convolutions. arXiv.
7. Kingma, D.P., and Welling, M. (2013). Auto-Encoding Variational Bayes. arXiv.
8. Provably secure generative steganography based on autoregressive model10.1007/978-3-030-11389-6_5 ↗
9. Chen, K., Zhou, H., Zhao, H., Chen, D., Zhang, W., and Yu, N. (2018). When Provably Secure Steganography Meets Generative Models. arXiv.
10. A novel image steganography method via deep convolutional generative adversarial networks10.1109/ACCESS.2018.2852771 ↗
11. Zhou, Z., Sun, H., Harit, R., Chen, X., and Sun, X. (2015, January 22–27). Coverless image steganography without embedding. Proceedings of the International Conference on Cloud Computing and Security, Geneva, Switzerland.10.1007/978-3-319-27051-7_11 ↗
12. Coverless information hiding based on bag-of-words model of image
13. Liu, J., Zhou, T., Zhang, Z., Ke, Y., Lei, Y.-Z., Zhang, M., and Yang, X. (2018, January 13–15). Digital cardan grille: A modern approach for information hiding. Proceedings of the International Conference on Computer Science and Artificial Intelligence, Shanghai, China.10.1145/3297156.3297255 ↗
14. Kishore, V., Chen, X., Wang, Y., Li, B., Weinberger, K.Q. (2022, January 25–29). Fixed neural network steganography: Train the images, not the network. Proceedings of the International Conference on Learning Representations, Virtually.
15. Zhong, Y., Liu, J., Luo, P., Ke, Y., and Cai, S. (2025). INR-Based Generative Steganography by Point Cloud Representation. arXiv.10.21203/rs.3.rs-5388114/v1 ↗
16. Implicit neural representation steganography by neuron pruning10.1007/s00530-024-01476-9 ↗
17. StegaINR4MIH: Steganography by implicit neural representation for multi-image hiding10.1117/1.JEI.33.6.063017 ↗
18. Liu, M., Ming, Z., Jun, L., Zhang, Y., and Ke, Y. (2017). Coverless information hiding based on generative adversarial networks. arXiv.
19. Liu, X., Ma, Z., Ma, J., Zhang, J., Schaefer, G., and Fang, H. (2022, January 19–24). Image Disentanglement Autoencoder for Steganography Without Embedding. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA.10.1109/CVPR52688.2022.00234 ↗
20. Volkhonskiy, D., Nazarov, I., Borisenko, B., and Burnaev, E. (2017, January 26–28). Steganographic Generative Adversarial Networks. Proceedings of the International Conference on Machine Vision, Dubai, United Arab Emirates.
21. Shi, H., Dong, J., Wang, W., Qian, Y., and Zhang, X. (2017). SSGAN: Secure Steganography Based on Generative Adversarial Networks. arXiv.10.1007/978-3-319-77380-3_51 ↗
22. Image steganography based on foreground object generation by generative adversarial networks in mobile edge computing with internet of things10.1109/ACCESS.2019.2913895 ↗
23. Park, H., Yoo, Y.J., and Kwak, N. (2018). MC-GAN: Multi-conditional generative adversarial network for image synthesis. arXiv.
24. LSB Matching revisited10.1109/LSP.2006.870357 ↗
25. Provably Secure Robust Image Steganography10.1109/TMM.2023.3330098 ↗
26. VAE-Stega: Linguistic Steganography Based on Variational Auto-Encoder10.1109/TIFS.2020.3023279 ↗
27. Zhu, J., Chen, Z., Yang, L., Xie, X., and Zhou, Y. (2024). Plug-and-Hide: Provable and Adjustable Diffusion Generative Steganography. arXiv.
28. Luo, Z., Li, S., Li, G., Qian, Z., and Zhang, X. (November, January 29). Securing Fixed Neural Network Steganography. Proceedings of the 31st ACM International Conference on Multimedia, Ottawa, ON, Canada.
29. Li, G., Li, S., Qian, Z., and Zhang, X. (November, January 28). Cover-separable Fixed Neural Network Steganography via Deep Generative Models. Proceedings of the 32nd ACM International Conference on Multimedia (MM’24), Melbourne, Australia.
30. Cheng, Y., Zhou, J., Chen, J., Yin, Z., and Zhang, X. (2025). RFNNS: Robust Fixed Neural Network Steganography with Popular Deep Generative Models. arXiv.
31. Han, S., Mao, H., and Dally, W.J. (2015). Deep Compression: Compressing Deep Neural Network with Pruning, Trained Quantization and Huffman Coding. arXiv.
32. Liu, Z., Sun, M., Zhou, T., Huang, G., and Darrell, T. (2018). Rethinking the Value of Network Pruning. arXiv.
33. Frankle, J., Dziugaite, G.K., Roy, D.M., and Carbin, M. (2020). Pruning Neural Networks at Initialization: Why are We Missing the Mark?. arXiv.
34. Park, D., Kim, S.H., Lee, S., and Kim, H.J. (2024). DDMI: Domain-Agnostic Latent Diffusion Models for Synthesizing High-Quality Implicit Neural Representations. arXiv.
35. Liu, J., Luo, P., and Ke, Y. (2023). Hiding Functions within Functions: Steganography by Implicit Neural Representations. arXiv.
36. Song, S., Yang, S., Yoo, C.D., and Kim, J. (October, January 30). Implicit Steganography Beyond the Constraints of Modality. Proceedings of the European Conference on Computer Vision, Krakow, Poland.
37. Image steganography based on generative implicit neural representation10.1117/1.JEI.33.6.063043 ↗
38. Dupont, E., Teh, Y.W., and Doucet, A. (2021, January 13–15). Generative Models as Distributions of Functions. Proceedings of the 24th International Conference on Artificial Intelligence and Statistics, Virtually.
39. Wei, P., Li, S., Zhang, X., Luo, G., Qian, Z., and Zhou, Q. (2022, January 10–14). Generative Steganography Network. Proceedings of the 30th ACM International Conference on Multimedia (MM’22), Lisbon, Portugal.10.1145/3503161.3548217 ↗
40. Zhuang, P., Abnar, S., Gu, J., Schwing, A., Susskind, J.M., and Bautista, M.A. (2023). Diffusion Probabilistic Fields. arXiv.
41. Du, Y., Collins, K.M., Tenenbaum, J.B., and Sitzmann, V. (2021, January 6–14). Learning Signal-Agnostic Manifolds of Neural Fields. Proceedings of the Neural Information Processing Systems, Online.
42. Mescheder, L., Geiger, A., and Nowozin, S. (2018, January 10–15). Which Training Methods for GANs do actually Converge?. Proceedings of the International Conference on Machine Learning, Stockholm, Sweden.
43. Wu, W., Qi, Z.A., and Li, F. (2019, January 16–20). PointConv: Deep Convolutional Networks on 3D Point Clouds. Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA.10.1109/CVPR.2019.00985 ↗
44. Karras, T., Aila, T., Laine, S., and Lehtinen, J. (May, January 30). Progressive Growing of GANs for Improved Quality, Stability, and Variation. Proceedings of the International Conference on Learning Representations (ICLR), Vancouver Convention Center, Vancouver, BC, Canada.
45. Karras, T., Laine, S., Aittala, M., Hellsten, J., Lehtinen, J., and Aila, T. (2020, January 14–19). Analyzing and Improving the Image Quality of StyleGAN. Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Online.10.1109/CVPR42600.2020.00813 ↗
46. Luo, Z., Guo, Q., Cheung, K.C., See, S., and Wan, R. (2023, January 2–6). CopyRNeRF: Protecting the CopyRight of Neural Radiance Fields. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), Paris, France.10.1109/ICCV51070.2023.02047 ↗
47. Song, Q., Luo, Z., Cheung, K.C., See, S., and Wan, R. (October, January 29). Protecting NeRFs’ Copyright via Plug-and-Play Watermarking Base Model. Proceedings of the European Conference on Computer Vision, Milan, Italy.
48. Boehm, B. (2014). StegExpose—A Tool for Detecting LSB Steganography. arXiv.