Towards a Better Training for Siamese CNNs on Kinship Verefication

  • Sellam AbdellahEmail author
  • Azzoune Hamid
Conference paper
Part of the Lecture Notes in Networks and Systems book series (LNNS, volume 64)


Kinship verification from facial images in the wild is a recent problem that received an increasing interest from the computer vision research community. Due to the limited size of the existing datasets, applying Deep Learning approaches results in a model that overfits to the training data, therefore, the purpose of this study is to reduce the degree of overfitting when training a Deep Learning model on kinship datasets. To this end, we propose a new training mechanism for siamese convnets, in which we train the model on all images from all types of kinship relations instead of training on each of these subsets separately, then we evaluate the model on each subset individually. Experimental results demonstrated that using this training method resulted in better performance compared to training on each subset separately, and allowed to achieve results comparable to the most recent state of the art approaches. This paper focuses on the impact of adding more data over adding the gender information by separating kinship relation types in different subsets.


Kinship verification Siamese CNN Deep learning Feature learning 


  1. 1.
    Fang, R., Tang, K.D., Snavely, N., Chen, T.: Towards computational models of kinship verification. In: 2010 17th IEEE International Conference on Image Processing (ICIP), pp. 1577–1580. IEEE (2010)Google Scholar
  2. 2.
    Zhou, X., Hu, J., Lu, J., Shang, Y., Guan, Y.: Kinship verification from facial images under uncontrolled conditions. In: Proceedings of the 19th ACM International Conference on Multimedia, pp. 953–956. ACM (2011)Google Scholar
  3. 3.
    Lu, J., Zhou, X., Tan, Y.P., Shang, Y., Zhou, J.: Neighborhood repulsed metric learning for kinship verification. IEEE Trans. Pattern Anal. Mach. Intell. 36(2), 331–345 (2014)CrossRefGoogle Scholar
  4. 4.
    Hu, J., Lu, J., Tan, Y.P.: Sharable and individual multi-view metric learning. IEEE Trans. Pattern Anal. Mach. Intell. (2017)Google Scholar
  5. 5.
    Hu, J., Lu, J., Tan, Y.P., Yuan, J., Zhou, J.: Local large-margin multi-metric learning for face and kinship verification. IEEE Trans. Circuits Syst. Video Technol. (2017)Google Scholar
  6. 6.
    Xu, M., Shang, Y.: Kinship measurement on face images by structured similarity fusion. IEEE Access 4, 10280–10287 (2016)CrossRefGoogle Scholar
  7. 7.
    Yan, H., Lu, J., Deng, W., Zhou, X.: Discriminative multimetric learning for kinship verification. IEEE Trans. Inf. Forensics Secur. 9(7), 1169–1178 (2014)CrossRefGoogle Scholar
  8. 8.
    Zhou, X., Shang, Y., Yan, H., Guo, G.: Ensemble similarity learning for kinship verification from facial images in the wild. Inf. Fusion 32, 40–48 (2016)CrossRefGoogle Scholar
  9. 9.
    Yan, H., Lu, J., Zhou, X.: Prototype-based discriminative feature learning for kinship verification. IEEE Trans. Cybern. 45(11), 2535–2545 (2015)CrossRefGoogle Scholar
  10. 10.
    Bottino, A., Vieira, T.F., Ul Islam, I.: Geometric and textural cues for automatic kinship verification. Int. J. Pattern Recognit. Artif. Intell. 29(03), 1556001 (2015)CrossRefGoogle Scholar
  11. 11.
    Zhou, X., Yan, H., Shang, Y.: Kinship verification from facial images by scalable similarity fusion. Neurocomputing 197, 136–142 (2016)CrossRefGoogle Scholar
  12. 12.
    Qin, X., Tan, X., Chen, S.: Mixed bi-subject kinship verification via multi-view multi-task learning. Neurocomputing 214, 350–357 (2016)CrossRefGoogle Scholar
  13. 13.
    Lan, R., Zhou, Y.: Quaternion-michelson descriptor for color image classification. IEEE Trans. Image Process. 25(11), 5281–5292 (2016)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Lan, R., Zhou, Y., Tang, Y.Y.: Quaternionic weber local descriptor of color images. IEEE Trans. Circuits Syst. Video Technol. 27(2), 261–274 (2017)CrossRefGoogle Scholar
  15. 15.
    Yan, H.: Kinship verification using neighborhood repulsed correlation metric learning. Image Vis. Comput. 60, 91–97 (2017)CrossRefGoogle Scholar
  16. 16.
    Patel, B., Maheshwari, R., Raman, B.: Evaluation of periocular features for kinship verification in the wild. Comput. Vis. Image Underst. 160, 24–35 (2017)CrossRefGoogle Scholar
  17. 17.
    Lu, J., Hu, J., Tan, Y.P.: Discriminative deep metric learning for face and kinship verification. IEEE Trans. Image Process. 26(9), 4269–4282 (2017)MathSciNetCrossRefGoogle Scholar
  18. 18.
    Zhang, K., Huang, Y., Song, C., Wu, H., Wang, L.: Kinship verification with deep convolutional neural networks. In: Proceedings of the British Machine Vision Conference (BMVC), pp. 148.1–148.12. BMVA Press (September 2015).
  19. 19.
    Bromley, J., Guyon, I., LeCun, Y., Säckinger, E., Shah, R.: Signature verification using a “siamese” time delay neural network. In: Advances in Neural Information Processing Systems, pp. 737–744 (1994)Google Scholar
  20. 20.
    Taigman, Y., Yang, M., Ranzato, M., Wolf, L.: Deepface: closing the gap to human-level performance in face verification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1701–1708 (2014)Google Scholar
  21. 21.
    Koch, G., Zemel, R., Salakhutdinov, R.: Siamese neural networks for one-shot image recognition. In: ICML Deep Learning Workshop. vol. 2 (2015)Google Scholar
  22. 22.
    Sellam, A., Azzoune, H.: All-subsets siamese convolutional neural network for kinship verification (2018).
  23. 23.
    Lu, J., Hu, J., Zhou, X., Shang, Y., Tan, Y.P., Wang, G.: Neighborhood repulsed metric learning for kinship verification. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2594–2601. IEEE (2012)Google Scholar
  24. 24.
    Chen, X., An, L., Yang, S., Wu, W.: Kinship verification in multi-linear coherent spaces. Multimed. Tools Appl. 76(3), 4105–4122 (2017)CrossRefGoogle Scholar
  25. 25.
    Dehshibi, M.M., Shanbehzadeh, J.: Cubic norm and kernel-based bi-directional PCA: toward age-aware facial kinship verification. Vis. Comput., 1–18 (2017)Google Scholar
  26. 26.
    Faraki, M., Harandi, M.T., Porikli, F.: No fuss metric learning, a hilbert space scenario. Pattern Recognit. Lett. 98, 83–89 (2017)CrossRefGoogle Scholar
  27. 27.
    Mahpod, S., Keller, Y.: Kinship verification using multiview hybrid distance learning. Comput. Vis. Image Underst. 167, 28–36 (2018)CrossRefGoogle Scholar
  28. 28.
    Yang, Y., Wu, Q.: A novel kinship verification method based on deep transfer learning and feature nonlinear mapping. DEStech Transactions on Computer Science and Engineering (aiea) (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Laboratory for Research in Artificial Intelligence, Computer Science DepartmentUniversity of Science and Technology Houari Boumediene (USTHB)El-Alia, Bab EzzouarAlgeria

Personalised recommendations