OmniGuard: Hybrid Manipulation Localization via Augmented Versatile Deep Image Watermarking

2024-12-02

arXiv.org

AI Summary

Existing watermarking methods for AIGC image editing suffer from inherent trade-offs among tampering localization accuracy, visual fidelity, and copyright extraction reliability, while also exhibiting fixed watermark placement and insufficient robustness. Method: We propose an enhanced adaptive watermarking framework integrating active embedding with passive blind extraction, enabling flexible watermark layout and high-precision local tampering localization. Our approach introduces a hybrid forensic architecture and a degradation-aware tampering extraction network, augmented by a lightweight AIGC-editing simulation layer to overcome positional rigidity. Results: Experiments demonstrate significant improvements over EditGuard: +4.25 dB in PSNR, +14.8% in F1-score under 20.7% noise, and +14.8% in average bit accuracy. The framework achieves superior visual fidelity, strong robustness against diverse distortions, and deployment flexibility—addressing critical authenticity and integrity challenges in AIGC-edited imagery.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Responsible Web: Algorithmic accountability and transparency on the web
• Search and Retrieval-Augmented AI: Agentic search

Abstract

With the rapid growth of generative AI and its widespread application in image editing, new risks have emerged regarding the authenticity and integrity of digital content. Existing versatile watermarking approaches suffer from trade-offs between tamper localization precision and visual quality. Constrained by the limited flexibility of previous framework, their localized watermark must remain fixed across all images. Under AIGC-editing, their copyright extraction accuracy is also unsatisfactory. To address these challenges, we propose OmniGuard, a novel augmented versatile watermarking approach that integrates proactive embedding with passive, blind extraction for robust copyright protection and tamper localization. OmniGuard employs a hybrid forensic framework that enables flexible localization watermark selection and introduces a degradation-aware tamper extraction network for precise localization under challenging conditions. Additionally, a lightweight AIGC-editing simulation layer is designed to enhance robustness across global and local editing. Extensive experiments show that OmniGuard achieves superior fidelity, robustness, and flexibility. Compared to the recent state-of-the-art approach EditGuard, our method outperforms it by 4.25dB in PSNR of the container image, 20.7% in F1-Score under noisy conditions, and 14.8% in average bit accuracy.

Problem

• Enhances digital content authenticity and integrity protection.
• Improves tamper localization precision and visual quality trade-offs.
• Boosts copyright extraction accuracy under AIGC-editing conditions.

Innovation

• Hybrid forensic framework for flexible watermark selection
• Degradation-aware tamper extraction network for precise localization
• Lightweight AIGC-editing simulation layer for enhanced robustness