Journal of Korea Society of Industrial Information Systems (한국산업정보학회논문지)

Korea Society of Industrial Information Systems (한국산업정보학회)
권호 표지
DOI QR코드

DOI QR Code

Deep Learning Based Fake Face Detection

딥 러닝 기반의 가짜 얼굴 검출

  • 김대희 (한밭대학교 제어계측공학과) ;
  • 최승완 (한밭대학교 제어계측공학과) ;
  • 곽수영 (한밭대학교 전자제어공학과)
  • Received : 2018.09.11
  • Accepted : 2018.10.15
  • Published : 2018.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, the increasing interest of biometric systems has led to the creation of many researches of biometrics forgery. In order to solve this forgery problem, this paper proposes a method of determining whether a synthesized face made of artificaial intelligence is real face or fake face. The proposed algorithm consists of two steps. Firstly, we create the fake face images using various GAN (Generative Adversarial Networks) algorithms. After that, deep learning algorithm can classify the real face image and the generated face image. The experimental results shows that the proposed algorithm can detect the fake face image which looks like the real face. Also, we obtained the classification accuracy of 88.7%.

최근 바이오인식 기술이 대중화됨에 따라 위 변조에 대응하는 연구 및 시도들이 많이 진행되고 있다. 본 논문에서 인공지능으로 만든 합성된 얼굴을 진짜 얼굴인지 합성된 가짜 얼굴인지를 판별하는 방법을 제안하고자 한다. 제안하는 알고리즘은 크게 2가지 단계로 구성되어 있다. 먼저, 실제 얼굴 사진에 여러 가지 GAN(Generative Adversarial Networks)알고리즘을 통해 합성된 가짜 얼굴을 생성하게 된다. 이후, 실제 얼굴 영상과 생성된 얼굴 영상을 딥러닝 알고리즘에 입력하여 진짜 또는 가짜인지 판별하도록 한다. 제안한 알고리즘은 실제 육안으로도 구별하기 어려운 합성 영상도 잘 구분하고, 테스트 결과 88.7%의 정확도를 확인하였다.

Keywords

References

  1. I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. C. Courville and Y. Bengio., "Generative Adversarial Nets," In Proceedings of NIPS, pp. 2672-2680, 2014.
  2. T. Karras, T. Aila S. Laine, J. Lehtinen., "Progressive Growing Of Gans For Improved Quality, Stability, And Variation," Published as a Conference Paper at ICLR, 2018.
  3. Supasorn Suwajanakorn, Steven M. Seitz and Ira Kemelmacher-Shlizerman., "Synthesizing Obama: Learning Lip Sync from Audio," Published University of Washington.
  4. J. Li, Y. Wang, T. Tan and A. K. Jain., "Live Face Detection Based on the Analysis of Fourier Spectra," In Biometric Technology for Human Identification, SPIE Vol. 5404, pp. 295-303, 2004.
  5. S. Lin, P. Gang and Zhaohui Wu., "Blinking-Based Live Face Detection Using Conditional Random Fields," Proc, AIB, pp. 252-260, 2007.
  6. Alec Radford, Luke Metz, Soumith Chintala., "Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks," arXiv preprint arXiv:1511.06434, 2015.
  7. Yoav Goldberg, Omer Levy., "Word2vec Explained: Deriving Mikolov et al.'s Negative-Sampling Word-Embedding Method," arXiv preprint arXiv:1402.3722, 2014.
  8. David Berthelot, Thomas Schumm, Luke Metz., "BEGAN: Boundary Equilibrium Generative Adversarial Networks," arXiv preprint arXiv:1703.10717, 2017.
  9. Diederik P Kingma, Max Welling., "Auto-Encoding Variational Bayes," arXiv preprint arXiv:1312.6114, 2013.
  10. Ziwei Liu, Ping Luo, Xiaogang Wang, Xiaoou Tang., "Large-Scale CelebFaces Attributes (CelebA) Dataset," Retrived from http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html.
  11. Andrew NG. "Autoencoders and Sparsity", Retrieved from http://ufldl.stanford.edu/wiki/index.php/Autoencoders_and_Sparsity.
  12. Ian Goodfellow. "Nips 2016 Tutorial: Generative Adversarial Networks," arXiv preprint arXiv:1701.00160, 2016.