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

2025-04-26

AI Summary

Existing screen-camera (SC) image watermarking approaches suffer from inaccurate noise modeling: mathematical models exhibit bias due to incomplete noise decomposition and neglect of coupled noise components, while supervised neural networks rely on scarce paired synthetic–real data and thus lack generalizability. Method: This paper introduces the Sim-to-Real paradigm to SC watermark robustness modeling for the first time, proposing an unsupervised, end-to-end differentiable noise layer. It aligns simulated and real SC noise distributions via unsupervised domain adaptation and adversarial feature constraints—without requiring paired data. Contribution/Results: Evaluated across diverse devices, illumination conditions, and viewing angles, our method achieves an average 12.7% improvement in watermark extraction accuracy and demonstrates significantly superior cross-device generalization compared to state-of-the-art methods.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements

Abstract

Unauthorized screen capturing and dissemination pose severe security threats such as data leakage and information theft. Several studies propose robust watermarking methods to track the copyright of Screen-Camera (SC) images, facilitating post-hoc certification against infringement. These techniques typically employ heuristic mathematical modeling or supervised neural network fitting as the noise layer, to enhance watermarking robustness against SC. However, both strategies cannot fundamentally achieve an effective approximation of SC noise. Mathematical simulation suffers from biased approximations due to the incomplete decomposition of the noise and the absence of interdependence among the noise components. Supervised networks require paired data to train the noise-fitting model, and it is difficult for the model to learn all the features of the noise. To address the above issues, we propose Simulation-to-Real (S2R). Specifically, an unsupervised noise layer employs unpaired data to learn the discrepancy between the modeling simulated noise distribution and the real-world SC noise distribution, rather than directly learning the mapping from sharp images to real-world images. Learning this transformation from simulation to reality is inherently simpler, as it primarily involves bridging the gap in noise distributions, instead of the complex task of reconstructing fine-grained image details. Extensive experimental results validate the efficacy of the proposed method, demonstrating superior watermark robustness and generalization compared to those of state-of-the-art methods.

Problem

• Unauthorized screen capturing causes security threats
• Existing noise layers fail to approximate SC noise effectively
• Proposing unsupervised noise layer to bridge simulation-reality gap

Innovation

• Unsupervised noise layer learns noise discrepancy
• Bridges gap between simulated and real noise
• Uses unpaired data for noise distribution learning