发表机构
Tianjin University of Technology; Shandong Artificial Intelligence Institute; Qilu University of Technology (Shandong Academy of Sciences)（天津理工大学; 山东人工智能研究院; 齐鲁工业大学（山东省科学院））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对深度伪造检测中跨分布泛化瓶颈，提出EISL框架，通过低秩投影解耦特征并设计环境干预模块，在多设置实验中提升了对未见伪造类型和环境变化的鲁棒性。
AI 中文摘要
跨分布泛化仍是深度伪造检测的关键瓶颈。尽管近期研究利用大规模视觉基础模型（Visual Foundation Models, VFMs）的语义先验，但存在一个值得注意却未充分探索的挑战：这些语义先验易受光照、风格等环境因素的干扰。关键在于，这种干扰会在伪造线索与环境模式之间建立虚假关联，严重限制了泛化能力。为应对这一根本挑战，我们提出了创新的环境不变子空间学习（Environment-Invariant Subspace Learning, EISL）框架。EISL的核心贡献在于，通过可学习的低秩投影将特征解耦为与伪造相关的正交不变因子和与环境相关的残差因子。为促进鲁棒的特征解耦，我们还设计了环境干预模块，生成多样化且具有挑战性的干预对，模拟分布外的环境变化，以引导模型发现真正不变的伪造表征。在跨数据集、跨生成器、全脸合成及损坏设置下的实验显示，与强大的检测器相比，我们的方法取得了一致的增益和有竞争力或领先的性能，证明其对未见伪造类型和环境变化的鲁棒性得到提升。本研究为理解和解决VFMs在深度伪造检测中的泛化障碍提供了新视角和有价值的探索。
英文摘要
Cross-distribution generalization remains a critical bottleneck in deepfake detection. While recent efforts leverage the semantic priors of large-scale visual foundation models (VFMs), a noteworthy yet underexplored challenge remains: the susceptibility of these semantic priors to environmental interference from factors such as lighting and style. Crucially, this interference establishes spurious correlations between forgery cues and environmental patterns that severely limit generalization. To address this fundamental challenge, we propose an innovative Environment-Invariant Subspace Learning (EISL) framework. The core contribution of EISL is that it aims to disentangle features into orthogonal forgery-relevant invariant factors and environment-related residual factors via a learnable low-rank projection. To facilitate robust feature disentanglement, we also design an Environmental Intervention module that generates diverse and challenging intervention pairs, simulating out-of-distribution environmental shifts to guide the model toward discovering truly invariant forgery representations. Experiments across cross-dataset, cross-generator, whole-face synthesis, and corruption settings show consistent gains and competitive or leading performance against strong detectors, demonstrating improved robustness to unseen forgery types and environmental variations. This work provides a new perspective and a valuable exploration for understanding and tackling the generalization barriers of VFMs in deepfake detection.
Comments12 pages, 4 figures, 11 tables
