Detection-of-Manipulated-Facial-Video

New York University Final project in ECE-GY 9123: Deep Learning

Team members: Huan Wang(hw2588) & Yunfan Xu(yx2277)

Introduction:

Manipulated videos, audio and images have been around for years, but the rise of deep learning and artificial intelligence have made them harder to detect. The misinformation created by Deepfakes, FaceSwap and Face2face can easily misdirect people and intensify social conflicts (Fridrich, J., & Kodovsky, J. (2012). Rich Models for Steganalysis of Digital Images. IEEE Transactions on Information Forensics and Security, 7(3), 868-882. doi:10.1109/tifs.2012.2190402). The spread of manipulated facial videos and images on social media networks encourages sociality to seriously consider its dangers and encourages researchers to develop powerful deepfake detection methods.