SAiW: Source-Attributable Invisible Watermarking for Proactive Deepfake Defense

2026-03-24

AI Summary

This work proposes an active invisible watermarking framework to combat the growing threat of deepfakes, which severely undermine information integrity and public trust. Addressing the limitations of existing detection methods—namely their lagging responsiveness and poor generalization—the framework embeds traceable source identity information directly during media generation, enabling origin verification and proactive forgery defense. The approach formulates watermark embedding as a source-conditional representation learning problem, supporting scalable multi-source watermark generation. A dual-purpose forensic decoder simultaneously reconstructs the watermark and performs source attribution. Identity information is injected via feature-level linear modulation, complemented by a perception-guided module informed by human visual system priors to ensure high imperceptibility and strong robustness. Experiments demonstrate that the framework achieves excellent visual quality and maintains resilience against compression, noise, geometric transformations, and adversarial perturbations across multiple deepfake datasets.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Data transparency and provenance
• Responsible Web: Mitigating misinformation and disinformation

Abstract

Deepfakes generated by modern generative models pose a serious threat to information integrity, digital identity, and public trust. Existing detection methods are largely reactive, attempting to identify manipulations after they occur and often failing to generalize across evolving generation techniques. This motivates the need for proactive mechanisms that secure media authenticity at the time of creation. In this work, we introduce SAiW, a Source-Attributed Invisible watermarking Framework for proactive deepfake defense and media provenance verification. Unlike conventional watermarking methods that treat watermark payloads as generic signals, SAiW formulates watermark embedding as a source-conditioned representation learning problem, where watermark identity encodes the originating source and modulates the embedding process to produce discriminative and traceable signatures. The framework integrates feature-wise linear modulation to inject source identity into the embedding network, enabling scalable multi-source watermark generation. A perceptual guidance module derived from human visual system priors ensures that watermark perturbations remain visually imperceptible while maintaining robustness. In addition, a dual-purpose forensic decoder simultaneously reconstructs the embedded watermark and performs source attribution, providing both automated verification and interpretable forensic evidence. Extensive experiments across multiple deepfake datasets demonstrate that SAiW achieves high perceptual quality while maintaining strong robustness against compression, filtering, noise, geometric transformations, and adversarial perturbations. By binding digital media to its origin through invisible yet verifiable markers, SAiW enables reliable authentication and source attribution, providing a scalable foundation for proactive deepfake defense and trustworthy media provenance.

Problem

• deepfake
• proactive defense
• media provenance
• source attribution
• invisible watermarking

Innovation

• source-attributable watermarking
• proactive deepfake defense
• invisible watermarking
• feature-wise linear modulation
• media provenance