Research map / Video / Video segmentation and tracking
Large-Capacity and Flexible Video Steganography via Invertible Neural Network
Chong Mou, Youmin Xu, Jiechong Song, Chen Zhao, Bernard Ghanem, Jian 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 |
|---|---|---|---|
| Increase the sliding window size from 1 to 5 (ablation) | LF-VSN · Vimeo-T200, 2, 4, and 6 videos | Stego and secret PSNR: 39.64 -> 41.08; 42.97 -> 44.43 (2 videos); 36.41 -> 37.86; 37.67 -> 40.42 (4 videos); 34.47 -> 35.96; 35.11 -> 39.97 (6 | Improved |
| Increase the number of invertible blocks from 12 to 20 (ablation) | LF-VSN · Vimeo-T200, 3 videos | Stego and secret PSNR: 38.03 -> 38.91; 41.99 -> 42.40 | Improved |
| Add frequency-domain concatenation using DWT instead of direct channel-wise concatenation (ablation) | LF-VSN · Vimeo-T200, 3 videos | Stego and secret PSNR: 36.85 -> 38.55; 40.36 -> 42.27 | Improved |
| Add the redundancy prediction module instead of random Gaussian sampling (ablation) | LF-VSN · Vimeo-T200, 3 videos | Stego and secret PSNR: 38.28 -> 38.55; 41.69 -> 42.27 | Improved |
| Add key-controllable recovery | LF-VSN · Vimeo-T200, 2, 4, and 6 videos | Stego and secret PSNR: 40.97/38.67; 44.24/41.04; 37.55/34.41; 40.21/37.15; 35.68/30.48; 36.94/31.95 | Hurt |
| Scale the number of hidden videos up to 7 | LF-VSN · Vimeo-T200, multiple videos | Stego and secret PSNR: 40.97 -> 35.01; 44.24 -> 35.71 (2 to 7 videos) | Mixed |
Closest papers by meaning
- Large-Capacity Image Steganography Based on Invertible Neural Networks (CVPR 2021 · similarity 0.92)
- HiNet: Deep Image Hiding by Invertible Network (ICCV 2021 · similarity 0.89)
- StegFormer: Rebuilding the Glory of Autoencoder-Based Steganography (AAAI 2024 · similarity 0.85)
- Purified and Unified Steganographic Network (CVPR 2024 · similarity 0.85)
- Efficient and Separate Authentication Image Steganography Network (ICML 2025 · similarity 0.84)
- Hiding Images in Diffusion Models by Editing Learned Score Functions (CVPR 2025 · similarity 0.83)
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.
