GANonymizer: Image Anonymization Method Integrating Object Detection and Generative Adversarial Network

Conference paper
Part of the EAI/Springer Innovations in Communication and Computing book series (EAISICC)


Sharing and analyzing image data from ubiquitous urban cameras must enable us to understand and predict various contexts of the city. Meanwhile, since such image data always contains privacy data such as people and cars, we cannot easily share and analyze the data through the Internet for the viewpoint of privacy protection. As a result, most of urban image data are only kept/shared within the camera owners, or even discarded to reduce risks of privacy data leakage. To solve the privacy problem and accelerate sharing of urban image data, we propose GANonymizer that automatically detects and removes objects related to privacy from the urban images. GANonymizer combines two neural networks: (1) a network which detects objects related to privacy such as persons and cars in an input image using object detection network and (2) a network that removes the detected objects naturally as though they are not exist originally. Through our experiment of applying GANonymizer to urban video images, we confirmed that GANonymizer partially achieved natural removal of objects related to privacy.


Privacy protection Urban image anonymization DNN 



Part of this research was supported by National Institute of Information and Communications Technology.


  1. 1.
    Acs, G., Melis, L., Castelluccia, C., De Cristofaro, E.: Differentially private mixture of generative neural networks. In: IEEE Transactions on Knowledge and Data Engineering (2018)Google Scholar
  2. 2.
    Barnes, C., Shechtman, E., Finkelstein, A., Goldman, D.B.: Patchmatch: a randomized correspondence algorithm for structural image editing. ACM Trans. Graph. 28(3), 24 (2009)CrossRefGoogle Scholar
  3. 3.
    Barnich, O., Van Droogenbroeck, M.: Vibe: a universal background subtraction algorithm for video sequences. IEEE Trans. Image Process. 20(6), 1709–1724 (2011)MathSciNetCrossRefGoogle Scholar
  4. 4.
    Chinomi, K., Nitta, N., Ito, Y., Babaguchi, N.: PriSurv: privacy protected video surveillance system using adaptive visual abstraction. In: International Conference on Multimedia Modeling, pp. 144–154. Springer, Berlin (2008)Google Scholar
  5. 5.
    Everingham, M., Van Gool, L., Williams, C.K., Winn, J., Zisserman, A.: The pascal visual object classes (voc) challenge. Int. J. Comput. Vis. 88(2), 303–338 (2010)CrossRefGoogle Scholar
  6. 6.
    Iizuka, S., Simo-Serra, E., Ishikawa, H.: Globally and locally consistent image completion. ACM Trans. Graph. 36(4), 107 (2017)CrossRefGoogle Scholar
  7. 7.
    Kawano, M., Mikami, K., Yokoyama, S., Yonezawa, T., Nakazawa, J.: Road marking blur detection with drive recorder. In: 2017 IEEE International Conference on Big Data (Big Data), pp. 4092–4097. IEEE, Piscataway (2017)Google Scholar
  8. 8.
    Kingma, D.P., Dhariwal, P.: Glow: generative flow with invertible 1x1 convolutions (2018). Preprint arXiv:1807.03039Google Scholar
  9. 9.
    Le, T., Almansa, A., Gousseau, Y., Masnou, S.: Motion-consistent video inpainting. In: ICIP 2017: IEEE International Conference on Image Processing (2017)Google Scholar
  10. 10.
    Liu, Y., Wu, L.: Geological disaster recognition on optical remote sensing images using deep learning. Procedia Comput. Sci. 91, 566–575 (2016)CrossRefGoogle Scholar
  11. 11.
    Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., Berg, A.C.: SSD: single shot multibox detector. In: European Conference on Computer Vision, pp. 21–37. Springer, Berlin (2016)Google Scholar
  12. 12.
    Maeda, H., Sekimoto, Y., Seto, T., Kashiyama, T., Omata, H.: Road damage detection using deep neural networks with images captured through a smartphone (2018). Preprint arXiv:1801.09454Google Scholar
  13. 13.
    Morishita, S., Maenaka, S., Nagata, D., Tamai, M., Yasumoto, K., Fukukura, T., Sato, K.: Sakurasensor: quasi-realtime cherry-lined roads detection through participatory video sensing by cars. In: Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, pp. 695–705. ACM, New York (2015)Google Scholar
  14. 14.
    Newson, A., Almansa, A., Fradet, M., Gousseau, Y., Pérez, P.: Video inpainting of complex scenes. SIAM J. Imag. Sci. 7(4), 1993–2019 (2014)MathSciNetCrossRefGoogle Scholar
  15. 15.
    Pathak, D., Krahenbuhl, P., Donahue, J., Darrell, T., Efros, A.A.: Context encoders: feature learning by inpainting. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2536–2544 (2016)Google Scholar
  16. 16.
    Ren, Z., Lee, Y.J., Ryoo, M.S.: Learning to anonymize faces for privacy preserving action detection (2018). Preprint arXiv:1803.11556Google Scholar
  17. 17.
    Shaikh, S.H., Saeed, K., Chaki, N.: Moving object detection using background subtraction. In: Moving Object Detection Using Background Subtraction, pp. 15–23. Springer, Cham (2014)Google Scholar
  18. 18.
    Shetty, R., Fritz, M., Schiele, B.: Adversarial scene editing: automatic object removal from weak supervision (2018). Preprint arXiv:1806.01911Google Scholar
  19. 19.
    Shu, J., Zheng, R., Hui, P.: Cardea: context-aware visual privacy protection from pervasive cameras (2016). Preprint arXiv:1610.00889Google Scholar
  20. 20.
    Wu, Z., Wang, Z., Wang, Z., Jin, H.: Towards privacy-preserving visual recognition via adversarial training: a pilot study (2018). Preprint arXiv:1807.08379Google Scholar
  21. 21.
    Yu, H., Lim, J., Kim, K., Lee, S.B.: Pinto: enabling video privacy for commodity IoT cameras. In: Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security, pp. 1089–1101. ACM, New York (2018)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Keio UniversityFujisawaJapan

Personalised recommendations