AI Paper Atlas

Research map / Image generation / Image restoration

Sim-to-Real: An Unsupervised Noise Layer for Screen-Camera Watermarking Robustness

Yufeng Wu, Xin Liao, Baowei Wang, Han Fang, Xiaoshuai Wu, Mingyue Chen, Guiling Wang

AAAI 2026 · Image generation › Image restoration · approach: Raw and low-light noise modeling · 3 citations (OpenAlex)

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 changedWhereMeasuredEffect
use S2R unsupervised noise layer instead of SSDSMCFN watermarking · S+L, 30 cm, 0 degBER (%): 5.1 -> 2.1Improved
use S2R noise layer instead of PIMoGMCFN watermarking · S+L, 30 cm, 40 degBER (%): 9.5 -> 6.0Improved
use S2R noise layer instead of StegaStampMCFN watermarking · PSNRPSNR (dB): 39.89 -> 42.27Improved
replace unsupervised S2R with supervised variant trained on 900 paired samples (ablation)S2R · S+L, 30 cm, 0 degBER (%): 1.6 -> 3.8Hurt
replace the MIMO-UNet unsupervised translation with CycleGAN (ablation)S2R · S+L, 30 cm, 0 degBER (%): 2.1 -> 2.9Hurt
replace the MIMO-UNet unsupervised translation with DualGAN (ablation)S2R · S+L, 30 cm, 0 degBER (%): 2.1 -> 3.5Hurt
add unsupervised refinement network on top of SSDS simulation (ablation)SSDS · S+L, 30 cm, 0 degBER (%): 5.0 vs 5.1 in Table 2Mixed
train S2R on a single device pair (I+E) vs combined SIM+LEA (ablation)S2R · S+L, 30 cm, 0 degBER (%): 2.1 -> 1.6Improved
test S2R on device pairs not used in S2R training (ablation)S2R · Cross-deviceBER (%)No clear effect
train S2R and watermark network on different datasets (ablation)S2R · Cross-sourceBER (%)No clear effect
increase shooting angle to 60 degreesS2R · S+LBER (%): 5.8 -> 10.2Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy