Research map / Image generation / Image restoration
HiNet: Deep Image Hiding by Invertible Network
Junpeng Jing, Xin Deng, Mai Xu, Jianyi Wang, Zhenyu Guan
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 |
|---|---|---|---|
| add wavelet transform (hide in wavelet domain) (ablation) | HiNet · cover/stego pair | PSNR (dB): 40.71 -> 44.23 | Improved |
| add wavelet transform (hide in wavelet domain) (ablation) | HiNet · secret/recovery pair | PSNR (dB): 44.22 -> 48.52 | Improved |
| add low-frequency wavelet loss Lfreq (ablation) | HiNet · SRNet detection | detection rate (%): 75.42 -> 55.86 | Improved |
| add low-frequency wavelet loss Lfreq (ablation) | HiNet · cover/stego pair | PSNR (dB): 44.23 -> 46.52 | Improved |
| add low-frequency wavelet loss Lfreq (ablation) | HiNet · secret/recovery pair | PSNR (dB): 48.52 -> 46.98 | Hurt |
| use dense blocks for rho, phi, eta instead of convolutional or residual blocks (ablation) | HiNet · cover/stego PSNR | PSNR (dB) | Improved |
| set lambda_freq to 10 (ablation) | HiNet · PSNR, SSIM and detection rate | PSNR (dB), SSIM, detection rate | Mixed |
| use invertible network with shared concealing/revealing parameters instead of two separate networks | Baluja · DIV2K cover/stego | PSNR (dB): 36.77 -> 48.99 | Improved |
| use invertible network with shared parameters instead of Weng et al. | Weng et al. · DIV2K secret/recovery | PSNR (dB): 38.93 -> 52.86 | Improved |
| use HiNet instead of Weng et al. for steganalysis resistance | Weng et al. · SRNet detection | detection accuracy (%): 75.03 -> 55.86 | Improved |
Later papers that compared with this method
21 later accepted papers report HiNet as a baseline in their results tables; 19 of them report a better number on most of the results they share with it.
- KIStego: Key-Independent Secure Image Distribution via Bipartite Structural Invariants (ICML 2026 · better on 2 of 2 shared results)
- SD-PSFNet: Sequential and Dynamic Point Spread Function Network for Image Deraining (AAAI 2026 · better on 2 of 2 shared results)
- Towards Illumination-Aware Restoration of Metalens-Captured Images: A New Dataset and a Strong Baseline (AAAI 2026 · better on 2 of 2 shared results)
- Anatomical Consistency and Adaptive Prior-informed Transformation for Multi-contrast MR Image Synthesis via Diffusion Model (CVPR 2025 · better on 6 of 6 shared results)
- SSHR: More Secure Generative Steganography with High-Quality Revealed Secret Images (ICML 2025 · better on 3 of 3 shared results)
- Robust Message Embedding via Attention Flow-Based Steganography (CVPR 2025 · better on 2 of 2 shared results)
- After the Party: Navigating the Mapping From Color to Ambient Lighting (ICCV 2025 · better on 1 of 1 shared result)
- StegFormer: Rebuilding the Glory of Autoencoder-Based Steganography (AAAI 2024 · better on 3 of 3 shared results)
13 more not listed.
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
- DeepMIH: Deep Invertible Network for Multiple Image Hiding (TPAMI 2022 · similarity 0.92)
- Large-Capacity Image Steganography Based on Invertible Neural Networks (CVPR 2021 · similarity 0.89)
- Hiding Images within Images (TPAMI 2019 · similarity 0.89)
- Purified and Unified Steganographic Network (CVPR 2024 · similarity 0.85)
- Hiding Images in Diffusion Models by Editing Learned Score Functions (CVPR 2025 · similarity 0.85)
- IWRN:A Robust Blind Watermarking Method for Artwork Image Copyright Protection Against Noise Attack (AAAI 2025 · similarity 0.84)
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.
