Collusion-Resistant Image-Agnostic Watermarking for Multi-Screen Shooting

2026-07-26

AI Summary

This work addresses the vulnerability of traditional copyright watermarks to collusion attacks in multi-screen capture scenarios, where high cross-screen similarity facilitates watermark removal. To counter this, the authors propose CoMSMark, the first image-agnostic watermarking framework specifically designed against multi-screen collusion. It generates screen-specific watermark residuals through a screen ID–driven style modulation mechanism and incorporates a collusion-suppression loss to attenuate shared components across screens. By integrating identity-aware embedding with high-entropy spoofing prediction, the method significantly enhances watermark uniqueness and security. Experiments demonstrate that CoMSMark achieves over 90% watermark detection accuracy under collusion removal attacks, while spoofing detection performance remains near random chance (~50%), and exhibits strong robustness across varying capture distances and angles.

Abstract

Screen-shooting poses a significant threat to confidential information protection. While existing screen-shooting watermarking methods enable copyright verification, the copyrighted images carrying the same copyright watermark across different screens often exhibit highly similar and estimable watermark patterns. These shared patterns can be exploited for watermark removal and forgery, a threat we term the multi-screen collusion attack. To mitigate this threat, we propose CoMSMark, a collusion-resistant image-agnostic watermarking framework for multi-screen shooting, which reduces shared residual components across screens to resist multi-screen collusion attacks. Specifically, we incorporate screen ID through a style modulation mechanism, enabling the encoder to generate screen-specific watermark residuals for reliable source attribution. We further introduce a collusion suppression loss that reduces shared residual components and encourages high-entropy predictions for forged samples, improving resistance to collusion attacks. Finally, to enable efficient large-scale distribution, CoMSMark employs an image-agnostic encoding paradigm that generates watermark residuals independently of image content. Extensive experiments demonstrate that CoMSMark effectively resists both collusion-based watermark removal and forgery. It maintains an average watermark accuracy above 90% under removal attacks while keeping forged-watermark accuracy near 50%. Moreover, CoMSMark achieves competitive robustness under diverse screen-shooting conditions, including varying capture distances and angles.

Problem

• multi-screen collusion attack
• watermark removal
• watermark forgery
• screen-shooting watermarking
• collusion-resistant

Innovation

• collusion-resistant
• image-agnostic watermarking
• multi-screen shooting
• style modulation
• watermark residual