This paper presents a method to automatically and efï¬?ciently detect face tampering in videos, and particularly focuses on two recent techniques used to generate hyperrealistic forged videos: Deepfake and Face2Face. Traditionalimageforensicstechniquesareusuallynotwellsuited to videos due to the compression that strongly degrades the data. Thus, this paper follows a deep learning approach and presents two networks, both with a low number of layers to focus on the mesoscopic properties of images. We evaluatethosefastnetworksonbothanexistingdatasetand a dataset we have constituted from online videos. The tests demonstrateaverysuccessfuldetectionratewithmorethan 98% for Deepfake and 95% for Face2Face.