DeepFake Detection for Human Face Images and Videos: A Comprehensive Survey

With the growing sophistication of deep learning and generative models, the creation of synthetic media such as DeepFakes has become increasingly convincing and widespread. DeepFakes pose serious threats across multiple sectors, from political misinformation to personal identity theft. This paper reviews the current progress in DeepFake detection techniques focused on human facial images and video content. It categorizes detection methodologies into feature-based approaches, deep learning models, biological signal analysis, and multimodal systems. Additionally, it discusses benchmark datasets, performance metrics, and the ongoing challenges faced by detection systems, such as poor generalization to unseen forgery methods. Finally, the survey outlines future directions essential for building more reliable and adaptable DeepFake detection solutions.