EXTRACTING DEEP LOCAL FEATURES TO DETECT MANIPULATED IMAGES OF HUMAN FACES

Recent developments in computer vision and machine learning have made it possible to create realistic manipulated videos of human faces, raising the issue of ensuring adequate protection against the malevolent effects unlocked by such capabilities. In this paper we propose that local image features which are shared across manipulated regions are a key element for the automatic detection of manipulated face images. We also design a lightweight architecture specifically for extracting such features and derive a multitask training scheme that consistently outperforms image class supervision alone. The trained networks achieve state-of-the-art results in the FaceForensics++ dataset using significantly reduced number of parameters and are shown to work well in detecting fully generated face images.