Research map / Image generation / Image restoration
Robust Image Steganography: Hiding Messages in Frequency Coefficients
Yuhang Lan, Fei Shang, Jianhua Yang, Xiangui Kang, Enping Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| embed messages in DCT coefficients rather than the spatial domain (ablation) | INN image steganography · JPEG QF 50 ablation | BER(%): 1.4 -> 0 | Improved |
| add discrimination loss Ld (ablation) | DCT-based INN steganography · Ablation | PSNR(dB): 43.29 -> 48.13 | Improved |
| add mutual information loss Lm and TWFM (ablation) | DCT-based INN steganography with Ld · Ablation | BER(%): 0.39 -> 0 | Improved |
| add mutual information loss Lm and TWFM (ablation) | DCT-based INN steganography with Ld · Ablation | PSNR(dB): 48.13 -> 48.41 | Improved |
| use DCT coefficients rather than spatial-domain embedding (ablation) | INN image steganography · JPEG QF 50 ablation | PSNR(dB): 36.75 -> 48.13 | Improved |
| use DCT embedding and the proposed Lm and TWFM designs | Robust image hiding · JPEG QF 10, unknown channel | BER: 33.17 -> 10.93 | Improved |
| use DCT embedding and the proposed Lm and TWFM designs | Robust image hiding · JPEG QF 50, unknown channel | BER: 1.35 -> 0.74 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- HiNet — HiNet: Deep Image Hiding by Invertible Network (ICCV 2021) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Large-Capacity Image Steganography Based on Invertible Neural Networks (CVPR 2021 · similarity 0.89)
- Robust Invertible Image Steganography (CVPR 2022 · similarity 0.89)
- RFNNS: Robust Fixed Neural Network Steganography with Universal Text-to-Image Models (AAAI 2026 · similarity 0.89)
- Robust Message Embedding via Attention Flow-Based Steganography (CVPR 2025 · similarity 0.88)
- Provably Secure Image Robust Steganography via Cross-modal Error Correction (AAAI 2025 · similarity 0.87)
- StegaStyleGAN: Towards Generic and Practical Generative Image Steganography (AAAI 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
