Research map / Image generation / Fast diffusion sampling
KIStego: Key-Independent Secure Image Distribution via Bipartite Structural Invariants
Ren Lijing, Denghui Zhang
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 |
|---|---|---|---|
| use bipartite shares with orthogonal Gaussian mapping instead of key-dependent ODE inversion | KIStego · UniStega recovery | PSNR: 23.9374 (DiffStega) -> 24.2491 | Improved |
| decouple secret from stego-image via shares | KIStego · UniStega concealment | ID similarity: 0.2255 (DiffStega) -> 0.0106 | Improved |
| use KIStego against steganalysis detectors | KIStego · UniStega steganalysis | detection accuracy: 0.582 (CRoSS) -> 0.502 on SRNet | Improved |
| anchor payload to discrete structural invariants under channel degradation | KIStego · UniStega JPEG Q=70 | PSNR: 21.16 (DiffStega) -> 23.14 | Improved |
| add differentiable sigmoid measurement surrogate instead of unguided latent DPS (ablation) | KIStego · Guidance ablation | PSNR: 18.56 -> 24.25 | Improved |
| replace straight-through error-diffusion gradient with differentiable surrogate (ablation) | KIStego · Guidance ablation | PSNR: 21.12 -> 24.25 | Improved |
| increase payload density to 4.0 BPP via spatial multiplexing of 8 shares (ablation) | KIStego · UniStega capacity | PSNR: 24.25 -> 20.15 | Hurt |
| apply measurement guidance every 10 steps (efficient mode) (ablation) | KIStego · Recovery latency | latency: approximately 30% lower latency | Improved |
| use modification-based embedding (HiNet) under channel degradation | HiNet · UniStega robustness | PSNR: 46.15 -> 10.26 under blur | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CRoSS — CRoSS: Diffusion Model Makes Controllable, Robust and Secure Image Steganography (NeurIPS 2023) · better on 6 of 6 shared results
- HiNet — HiNet: Deep Image Hiding by Invertible Network (ICCV 2021) · better on 2 of 2 shared results
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
- SSHR: More Secure Generative Steganography with High-Quality Revealed Secret Images (ICML 2025 · similarity 0.86)
- Provably Secure Image Robust Steganography via Cross-modal Error Correction (AAAI 2025 · similarity 0.85)
- SecureGS: Boosting the Security and Fidelity of 3D Gaussian Splatting Steganography (ICLR 2025 · similarity 0.85)
- StegaStyleGAN: Towards Generic and Practical Generative Image Steganography (AAAI 2024 · similarity 0.84)
- StegoGAN: Leveraging Steganography for Non-Bijective Image-to-Image Translation (CVPR 2024 · similarity 0.83)
- Secure Seed-Based Multi-bit Watermarking for Diffusion Models from First Principles (NeurIPS 2026 · similarity 0.82)
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.
