无需伪造先验的伪造感知唇部表征学习
Learning Forgery-Aware Lip Representations Without Forgery Priors
提出仅用真实数据训练的伪造检测范式：用非对称对比目标把真实样本压紧、把潜在伪造推远，并用各向同性高斯正则约束真实表征空间，可直接迁移到其他「无伪造先验」的深度伪造检测与开放集异常识别任务。
原文依据 an asymmetric contrastive objective that compacts real samples while repelling potential forgeries in representation space
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论视觉说话人认证（VSA）通过分析提示语音期间的唇部动态来验证身份，相比全脸方法具有更强的隐私性，同时保持高安全应用所需的判别能力。然而，近期说话人脸生成（TFG）的进展使得逼真的伪造能够紧密模仿与语音同步的唇部动态，对 VSA 系统构成严重威胁。主流防御方法严重依赖通过经验风险最小化在已知伪造上训练的有监督分类器，导致对未见攻击的泛化能力差、依赖不断更新的伪造数据，并且在缺乏有效伪造先验时完全失效。本文重新审视伪造检测器的设计，认为对伪造先验的过度依赖阻碍了对真实视频中固有丰富真实性信号的利用。我们提出一种仅用真实数据训练的新型检测器，通过三个关键组件学习伪造感知表征：（1）从真实视频中捕获伪造指示统计量的轻量模块；（2）非对称对比目标，在表征空间中压缩真实样本同时排斥潜在伪造；（3）有理论依据的正则项，将真实表征塑造成可处理的各向同性高斯分布。为支持严格评估，我们引入涵盖多种 TFG 伪造的基准套件。在八种现代伪造攻击和十种最优（SOTA）检测器上，我们的方法在错误率上降低超过 10%，同时以极小开销保持身份验证能力，并在更贴近真实世界场景的数据集上表现出一致增益。
英文原文摘要 1737 字符
Visual Speaker Authentication (VSA) verifies identity by analyzing lip dynamics during prompted speech, offering enhanced privacy compared to full-face methods while maintaining discriminability for high-security applications. However, recent advances in talking face generation (TFG) have enabled realistic forgeries that closely mimic lip dynamics in sync with speech, posing severe threats to VSA systems. Prevailing defenses rely heavily on supervised classifiers trained on known forgeries via empirical risk minimization, resulting in poor generalization to unseen attacks, dependency on continuously updated fake data, and complete failure in the absence of effective forgery priors. In this paper, we revisit the design of forgery detectors and argue that over-reliance on fake priors hinders the exploitation of rich authenticity signals inherently present in real videos. We propose a novel detector trained exclusively on authentic data, learning forgery-aware representations through three key components: (1) lightweight modules that capture forgery-indicative statistics from real videos; (2) an asymmetric contrastive objective that compacts real samples while repelling potential forgeries in representation space; and (3) a theoretically grounded regularizer that shapes real representations into a tractable, isotropic Gaussian. To support rigorous evaluation, we introduce a benchmark suite spanning diverse TFG forgeries. Across eight modern forgery attacks and ten state-of-the-art (SOTA) detectors, our method achieves over a 10% reduction in error rates while preserving identity-verification capability with minimal overhead, and demonstrates consistent gains on datasets that better emulate real-world scenarios.