用小波与图网络实现抗攻击的深度伪造水印，保护隐私不被滥用。
WaveGuard: Robust Deepfake Detection and Source Tracing via Dual-Tree Complex Wavelet and Graph Neural Networks
- 在高频子带嵌入水印，利用双树复小波变换增强隐蔽性
- 通过图神经网络保持视觉质量，抗压缩、裁剪等攻击性能更强
- 适合需要可追溯身份的AI内容安全场景
深度伪造技术日益威胁隐私与身份安全。为此，我们提出WaveGuard，一种主动水印框架，通过频域嵌入和基于图的结构一致性提升鲁棒性与不可感知性。具体而言，采用双树复小波变换（DT-CWT）将水印嵌入高频子带，并设计结构一致性图神经网络（SC-GNN）以维持视觉质量。同时引入注意力模块提升嵌入精度。在人脸替换与重演任务上的实验表明，WaveGuard在鲁棒性和视觉质量方面均优于现有先进方法。代码已开源：https://github.com/vpsg-research/WaveGuard。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。