Embedded AutoEncoders: A Novel Framework for Face De-identification

  • Jianqi Liu
  • Jun Liu
  • Pan LiEmail author
  • Zhengze Kuang
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 1227)


Recent advances in deep learning and big data have greatly promoted the development of image recognition technology. In the meantime, however, it also makes it more challenging to protect human identify information. In this paper, we propose a novel framework called Embedded AutoEncoders to address face de-identification problem in deep learning. The structure of our framework contains two parts: a Privacy Removal Network and a Feature Selection Network. The main objective of our framework is to ensure that the Privacy Removal Network is capable of discarding information involving privacy and retaining desired information for certain image recognition applications. In order to achieve this goal, the design of the Privacy Removal Network is crucial. Specifically, we employ two different autoencoders, one of which is embedded within the other. We evaluate the proposed framework through extensive experiments, which show that the Embedded AutoEncoders framework can not only effectively retain data utility, but also protect personal identity information.


Deep learning Autoencoders Image recognition Face de-identification Privacy 


  1. 1.
    Boyle, M., Edwards, C., Greenberg, S.: The effects of filtered video on awareness and privacy. In: Proceedings of the 2000 ACM Conference on Computer Supported Cooperative Work, pp. 1–10, December 2000Google Scholar
  2. 2.
    Ribaric, S., Pavesic, N.: An overview of face de-identification in still images and videos. In: 2015 11th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG), vol. 4, pp. 1–6. IEEE, May 2015Google Scholar
  3. 3.
    Newton, E.M., Sweeney, L., Malin, B.: Preserving privacy by de-identifying face images. IEEE Trans. Knowl. Data Eng. 17(2), 232–243 (2005)CrossRefGoogle Scholar
  4. 4.
    Sweeney, L.: k-anonymity: a model for protecting privacy. Int. J. Uncertainty Fuzziness Knowl.-Based Syst. 10(05), 557–570 (2002)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Gross, R., Sweeney, L., De la Torre, F., Baker, S.: Model-based face de-identification. In: 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW 2006), p. 161. IEEE, June 2006Google Scholar
  6. 6.
    Gross, R., Airoldi, E., Malin, B., Sweeney, L.: Integrating utility into face de-identification. In: Danezis, G., Martin, D. (eds.) PET 2005. LNCS, vol. 3856, pp. 227–242. Springer, Heidelberg (2006). Scholar
  7. 7.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)Google Scholar
  8. 8.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
  9. 9.
    Szegedy, C., et al.: Going deeper with convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1–9 (2015)Google Scholar
  10. 10.
    Osia, S.A., et al.: A hybrid deep learning architecture for privacy-preserving mobile analytics. IEEE Internet Things J. 7, 4505–4518 (2020)CrossRefGoogle Scholar
  11. 11.
    Chi, H., Hu, Y.H.: Face de-identification using facial identity preserving features. In: 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP), pp. 586–590. IEEE, December 2015Google Scholar
  12. 12.
    Malekzadeh, M., Clegg, R.G., Haddadi, H.: Replacement autoencoder: a privacy-preserving algorithm for sensory data analysis. In: 2018 IEEE/ACM Third International Conference on Internet-of-Things Design and Implementation (IoTDI), pp. 165–176. IEEE (2018)Google Scholar
  13. 13.
    Mirjalili, V., Raschka, S., Namboodiri, A., Ross, A.: Semi-adversarial networks: convolutional autoencoders for imparting privacy to face images. In: 2018 International Conference on Biometrics (ICB), pp. 82–89. IEEE, February 2018Google Scholar
  14. 14.
    Vincent, P., Larochelle, H., Bengio, Y., Manzagol, P.A.: Extracting and composing robust features with denoising autoencoders. In: Proceedings of the 25th International Conference on Machine Learning, pp. 1096–1103, July 2008Google Scholar
  15. 15.
    Wu, Y., Yang, F., Ling, H.: Privacy-protective-GAN for face de-identification. arXiv preprint arXiv:1806.08906 (2018)
  16. 16.
    Siwek, K., Osowski, S.: Autoencoder versus PCA in face recognition. In: 2017 18th International Conference on Computational Problems of Electrical Engineering (CPEE), pp. 1–4. IEEE, September 2017Google Scholar
  17. 17.
    Almotiri, J., Elleithy, K., Elleithy, A.: Comparison of autoencoder and Principal Component Analysis followed by neural network for e-learning using handwritten recognition. In: 2017 IEEE Long Island Systems, Applications and Technology Conference (LISAT), pp. 1–5. IEEE, May 2017Google Scholar
  18. 18.
    Hornik, K., Stinchcombe, M., White, H.: Multilayer feedforward networks are universal approximators. Neural Netw. 2(5), 359–366 (1989)CrossRefGoogle Scholar
  19. 19.
    Bengio, Y., Courville, A., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35(8), 1798–1828 (2013)CrossRefGoogle Scholar
  20. 20.
    Liu, Z., Luo, P., Wang, X., Tang, X.: Deep learning face attributes in the wild. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 3730–3738 (2015)Google Scholar
  21. 21.
    Parkhi, O.M., Vedaldi, A., Zisserman, A.: Deep face recognition. In: BMVC, vol. 1, no. 3, p. 6 (2015)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.University of Electronic Science and Technology of ChinaChengduChina

Personalised recommendations