Research map / Image generation / Image restoration
Large-Capacity Image Steganography Based on Invertible Neural Networks
Shao-Ping Lu, Rong Wang, Tao Zhong, Paul L. Rosin
Check 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 the shared-parameter invertible ISN instead of the separate-network method [5] | image steganography · ImageNet and Paris StreetView, hide 1 image | container and revealed-image PSNR: 36.02 -> 38.05; 32.75 -> 35.38; 36.80 -> 40.49; 39.03 -> 43.33 | Improved |
| Use the shared-parameter invertible ISN instead of the separate-network method [5] | image steganography · ImageNet and Paris StreetView, hide 2 images | container and revealed-image PSNR: 30.18 -> 36.86; 29.17 -> 32.21; 37.14 -> 39.14; 34.73 -> 39.05 | Improved |
| Increase container-loss weight αco (ablation) | ISN · Paris StreetView, hide 1 or 2 images | container PSNR: 27.64 -> 40.49; 27.08 -> 39.14 | Improved |
| Increase container-loss weight αco (ablation) | ISN · Paris StreetView, hide 1 image | revealed-image PSNR: 41.38; 42.26; 41.16; 41.99; 43.33; 40.73 | Mixed |
| Increase container-loss weight αco (ablation) | ISN · Paris StreetView, hide 2 images | revealed-image PSNR: 38.04; 38.15; 37.48; 39.26; 38.87; 39.05 | Mixed |
| Use 16 rather than 8 invertible blocks (ablation) | ISN · Paris StreetView, hide 4 images | revealed-image PSNR: 31.12 -> 33.02 | Improved |
| Use 16 rather than 8 invertible blocks (ablation) | ISN · Paris StreetView, hide 4 images | container PSNR: 36.53 -> 36.03 | Hurt |
Closest papers by meaning
- Large-Capacity and Flexible Video Steganography via Invertible Neural Network (CVPR 2023 · similarity 0.92)
- HiNet: Deep Image Hiding by Invertible Network (ICCV 2021 · similarity 0.89)
- Robust Image Steganography: Hiding Messages in Frequency Coefficients (AAAI 2023 · similarity 0.89)
- StegFormer: Rebuilding the Glory of Autoencoder-Based Steganography (AAAI 2024 · similarity 0.86)
- Purified and Unified Steganographic Network (CVPR 2024 · similarity 0.85)
- Robust Message Embedding via Attention Flow-Based Steganography (CVPR 2025 · 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.
