Deepfake Detection using Supervised Temporal Feature Extraction model and LSTM

지도 학습한 시계열적 특징 추출 모델과 LSTM을 활용한 딥페이크 판별 방법

  • Lee, Chunghwan (Department of Electronic Engineering, Hanyang University) ;
  • Kim, Jaihoon (Department of Electronic Engineering, Hanyang University) ;
  • Yoon, Kijung (Department of Electronic Engineering, Hanyang University)
  • 이정환 (한양대학교 융합전자공학부) ;
  • 김재훈 (한양대학교 융합전자공학부) ;
  • 윤기중 (한양대학교 융합전자공학부)
  • Published : 2021.11.26

Abstract

As deep learning technologies becoming developed, realistic fake videos synthesized by deep learning models called "Deepfake" videos became even more difficult to distinguish from original videos. As fake news or Deepfake blackmailing are causing confusion and serious problems, this paper suggests a novel model detecting Deepfake videos. We chose Residual Convolutional Neural Network (Resnet50) as an extraction model and Long Short-Term Memory (LSTM) which is a form of Recurrent Neural Network (RNN) as a classification model. We adopted cosine similarity with hinge loss to train our extraction model in embedding the features of Deepfake and original video. The result in this paper demonstrates that temporal features in the videos are essential for detecting Deepfake videos.

Keywords