提升视频伪造检测的公平性与可解释性，避免对不同人群的偏见。
Fair and Interpretable Deepfake Detection in Videos
- 结合时序建模与人口统计感知增强，提升检测鲁棒性。
- 在多个数据集上实现最佳公平性与准确率平衡，优于现有方法。
- 通过概念提取使决策过程透明，适合非专家理解结果。
现有深度伪造检测方法常存在偏见、缺乏透明度且未能有效捕捉时序信息，导致在不同人口群体中决策偏差和结果不可靠。本文提出一种面向公平性的深度伪造检测框架，融合时序特征学习与人口统计感知的数据增强，以提升公平性与可解释性。方法采用基于序列的聚类进行时序建模，结合概念提取提升检测可靠性，并支持非专业用户理解判断依据。此外，引入人口统计感知的数据增强策略，平衡低频群体样本，并使用频域变换保留伪造痕迹，缓解偏见并增强泛化能力。在FaceForensics++、DFD、Celeb-DF和DFDC等多个数据集上，采用Xception、ResNet等先进架构的大量实验表明，所提方法在公平性与准确率之间实现了最优权衡，优于当前主流方法。
原文摘要 · Abstract (English)
Existing deepfake detection methods often exhibit bias, lack transparency, and fail to capture temporal information, leading to biased decisions and unreliable results across different demographic groups. In this paper, we propose a fairness-aware deepfake detection framework that integrates temporal feature learning and demographic-aware data augmentation to enhance fairness and interpretability. Our method leverages sequence-based clustering for temporal modeling of deepfake videos and concept extraction to improve detection reliability while also facilitating interpretable decisions for non-expert users. Additionally, we introduce a demography-aware data augmentation method that balances underrepresented groups and applies frequency-domain transformations to preserve deepfake artifacts, thereby mitigating bias and improving generalization. Extensive experiments on FaceForensics++, DFD, Celeb-DF, and DFDC datasets using state-of-the-art (SoTA) architectures (Xception, ResNet) demonstrate the efficacy of the proposed method in obtaining the best tradeoff between fairness and accuracy when compared to SoTA.
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。