Evidence-based Decision Modeling for Synthetic Face Detection with Uncertainty-driven Active Learning
2026-05-11 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionCryptography and Security
AI summaryⓘ
The authors address the problem of detecting fake facial images, which is important because these images can be used illegally. They point out that current detection methods are often too confident and struggle with unfamiliar images, partly because they depend on a technique called Softmax. To fix this, the authors propose a new method called EMSFD that estimates uncertainty explicitly and uses this uncertainty to choose which images to label, reducing the need for lots of labeled data. Their experiments show that EMSFD improves detection accuracy by about 15% and makes the system more reliable and understandable.
deep generative modelssynthetic face detectionSoftmax activationOut-of-Distribution (OOD)uncertainty estimationDirichlet distributionactive learningmodel generalizationlabeling costinterpretability
Authors
Qingchao Jiang, Zhenxuan Hou, Zhiying Zhu, Zhenxing Qian, Xinpeng Zhang, Zaiwang Gu
Abstract
With the rapid development of deep generative models, forged facial images are massively exploited for illegal activities. Although existing synthetic face detection methods have achieved significant progress, they suffer from the inherent limitation of overconfidence due to their reliance on the Softmax activation function. Thus, these methods often lead to unreliable predictions when encountering unknown Out-of-Distribution (OOD) images, and cannot ascertain the model's uncertainty in its prediction. Meanwhile, most existing methods require massive high-quality annotated data, which greatly limits their practicability across diverse scenarios. To address these limitations, we propose EMSFD (Evidence-based decision Modeling for Synthetic Face Detection with uncertainty-driven active learning), an approach designed to enhance detection reliability and generalizability. Specifically, EMSFD models class evidence using the Dirichlet distribution and explicitly incorporates model uncertainty into the prediction process. Furthermore, during training, the estimated uncertainty is exploited to prioritize more informative samples from the unlabeled pool for annotation, thereby reducing labeling cost and improving model generalization. Extensive experimental evaluations demonstrate that our method enhances the interpretability of synthetic face detection. Meanwhile, our method yields a 15\% increase in accuracy compared to existing state-of-the-art (SOTA) baselines, which demonstrates the superior detection performance and generalizability of our approach. Our code is available at: https://github.com/hzx111621/EMSFD.