用轻量设计提升深度伪造检测跨域泛化能力
Patch-Discontinuity Mining for Generalized Deepfake Detection
- 基于大模型迁移，设计简洁网络捕捉真假人脸差异特征
- 在跨域与跨篡改场景下性能领先，仅需0.28万参数
- 无需增加参数即可增强泛化，适合实际部署
生成式人工智能的快速发展使得高度逼真的虚假人脸图像得以生成，对个人隐私和网络信息完整性构成严重威胁。现有深度伪造检测方法多依赖手工设计的取证线索和复杂架构，在同域设置下表现良好，但在面对未见伪造模式时性能显著下降。本文提出GenDF框架，通过将强大的大规模视觉模型迁移至深度伪造检测任务，采用紧凑简洁的网络设计。GenDF引入深度伪造特异性表征学习以捕捉真实与虚假人脸间的判别性模式，通过特征空间重分布缓解分布不匹配问题，并采用分类不变特征增强策略提升泛化能力，且不增加额外可训练参数。大量实验表明，GenDF在跨域和跨篡改设置下均达到最先进泛化性能，仅需0.28M可训练参数，验证了该框架的有效性与高效性。
原文摘要 · Abstract (English)
The rapid advancement of generative artificial intelligence has enabled the creation of highly realistic fake facial images, posing serious threats to personal privacy and the integrity of online information. Existing deepfake detection methods often rely on handcrafted forensic cues and complex architectures, achieving strong performance in intra-domain settings but suffering significant degradation when confronted with unseen forgery patterns. In this paper, we propose GenDF, a simple yet effective framework that transfers a powerful large-scale vision model to the deepfake detection task with a compact and neat network design. GenDF incorporates deepfake-specific representation learning to capture discriminative patterns between real and fake facial images, feature space redistribution to mitigate distribution mismatch, and a classification-invariant feature augmentation strategy to enhance generalization without introducing additional trainable parameters. Extensive experiments demonstrate that GenDF achieves state-of-the-art generalization performance in cross-domain and cross-manipulation settings while requiring only 0.28M trainable parameters, validating the effectiveness and efficiency of the proposed framework.
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。