In daily life, unique biological information is often used in identity authentication, such as facial images, fingerprints, iris, and voiceprints, etc. Compared with other biological information, facial images have the following advantages: lower acquisition cost, more information, and no physical contact. Therefore, facial images have been extensively utilized in identification and authentication services. New applications, such as face payment, face retrieval, and face check-in have emerged, entering daily life in an all-round way; the "face brushing era" has arrived.

However, the security of these applications is threatened with the rapid development of digital image editing technology and software. There is not only traditional software like Photoshop, but also some emerging software such as Face2Face, FaceSwap, DeepFake, and so on, which can create realistic faces or modify facial details. Then, even a non-professional person can easily generate high-quality facial images to carry out various attacks, such as presentation attacks, morphing attacks, adversarial attacks, and so on. The research on facial image forensics is of great significance for fighting against illegal and criminal activities.

This Special Issue will provide a significant collective contribution to the field of face image forensics and security and focuses on data-driven approaches using machine learning. Original research and review articles are welcome.

Potential topics include but are not limited to the following:

• GAN-generated face detection
• Generation of face adversarial examples
• Face adversarial example recovery
• Digital face manipulation detection and localization
• Facial identity swap forensics
• Facial synthesis forensics
• Facial attribute manipulation forensics
• Facial expression swap forensics
• Facial morphing attack forensics
• Face liveness detection
• Face watermarking