WaveGuard: Robust Deepfake Detection and Source Tracing via Dual-Tree Complex Wavelet and Graph Neural Networks

TL;DR

WaveGuard introduces a novel watermarking framework using frequency-domain embedding and graph neural networks to improve deepfake detection robustness and visual quality.

Contribution

It combines DT-CWT and SC-GNN with an attention module for enhanced watermark robustness and imperceptibility in deepfake detection.

Findings

Outperforms state-of-the-art methods in robustness.

Maintains high visual quality in watermarked images.

Effective in face swap and reenactment tasks.

Abstract

Deepfake technology poses increasing risks such as privacy invasion and identity theft. To address these threats, we propose WaveGuard, a proactive watermarking framework that enhances robustness and imperceptibility via frequency-domain embedding and graph-based structural consistency. Specifically, we embed watermarks into high-frequency sub-bands using Dual-Tree Complex Wavelet Transform (DT-CWT) and employ a Structural Consistency Graph Neural Network (SC-GNN) to preserve visual quality. We also design an attention module to refine embedding precision. Experimental results on face swap and reenactment tasks demonstrate that WaveGuard outperforms state-of-the-art methods in both robustness and visual quality. Code is available at https://github.com/vpsg-research/WaveGuard.