access icon free Neural class-specific regression for face verification

© The Institution of Engineering and Technology
Received 09/05/2017, Accepted 07/09/2017, Revised 02/08/2017, Published 26/09/2017

Face verification is a problem approached in the literature mainly using non-linear class-specific subspace learning techniques. While it has been shown that kernel-based class-specific discriminant analysis is able to provide excellent performance in small- and medium-scale face verification problems, its application in today's large-scale problems is difficult due to its training space and computational requirements. In this study, generalising on kernel-based class-specific discriminant analysis, it is shown that class-specific subspace learning can be cast as a regression problem. This allows them to derive linear, (reduced) kernel and neural network-based class-specific discriminant analysis methods using efficient batch and/or iterative training schemes, suited for large-scale learning problems. The authors test the performance of these methods in two datasets describing medium- and large-scale face verification problems.

Inspec keywords: learning (artificial intelligence); face recognition; regression analysis; neural nets

Other keywords: iterative training schemes; neural network-based class-specific discriminant analysis methods; regression problem; class-specific subspace learning; linear-based class-specific discriminant analysis methods; kernel-based class-specific discriminant analysis; batch training schemes; neural class-specific regression; medium-scale face verification problems; large-scale face verification problems; large-scale learning problems

Subjects: Other topics in statistics; Neural computing techniques; Image recognition; Computer vision and image processing techniques; Other topics in statistics

References

    1. 1)
      • 18. Lee, Y., Huang, S.: ‘Reduced support vector machines: a statistical theory’, IEEE Trans. Neur. Netw., 2007, 18, (1), pp. 1–13.
    2. 2)
      • 10. Baudat, G., Anouar, F.: ‘Generalized discriminant analysis using a kernel approach’, Neural Comput., 2000, 12, pp. 2385–2404.
    3. 3)
      • 5. Duda, R., Hart, P., Stork, D.: ‘Pattern classification’ (Wiley-Interscience, 2000, 2nd edn.).
    4. 4)
      • 17. Drineas, P., Mahoney, M.: ‘On the Nyström method for approximating a gram matrix for improved kernel-based learning’, J. Mach. Learn. Res., 2005, 6, pp. 2153–2275.
    5. 5)
      • 7. Ye, J.: ‘Least squares linear discriminant analysis’. In Proceedings of the 24th international conference on Machine learning, June 2007, (ACM) pp. 1087–1093.
    6. 6)
      • 13. Goudelis, G., Zafeiriou, S., Tefas, A., et al: ‘Class-specific kernel discriminant analysis for face verification’, IEEE Trans. Inf. Forensics Sec., 2007, 2, (3), pp. 570–587.
    7. 7)
      • 28. Tang, E.K., Suganthan, P.N., Yao, X., et al: ‘Linear dimensionality reduction using relevance weighted LDA’, Pattern Recognit., 2005, 38, (4), pp. 485–493.
    8. 8)
      • 26. Schölkopf, B., Smola, A.J.: ‘Learning with kernels: support vector machines, regularization, optimization, and beyond’ (MIT Press, 2001), ISBN: 9780262253437.
    9. 9)
      • 21. Iosifidis, A., Gabbouj, M.: ‘Prototype-based class-specific nonlinear subspace learning for large-scale face verification’. Int. Conf. Image Processing Theory, Tools and Applications, 2016.
    10. 10)
      • 9. Iosifidis, A., Tefas, A., Pitas, I.: ‘On the optimal class representation in linear discriminant analysis’, IEEE Trans. Neural Netw. Learn. Syst., 2003, 24, (9), pp. 1491–1497.
    11. 11)
      • 1. Barr, P., Noble, J., Biddle, R.: ‘Video game values: human-computer interaction and games’, Interact. Comput., 2011, 19, (2), pp. 180–195.
    12. 12)
      • 3. Iosifidis, A., Tefas, A., Pitas, I.: ‘Activity based person identification using fuzzy representation and discriminant learning’, IEEE Trans. Inf. Forensics Sec., 2012, 7, (2), pp. 530–542.
    13. 13)
      • 19. Iosifidis, A., Gabbouj, M.: ‘Scaling up class-specific kernel discriminant analysis for large-scale face verification’, IEEE Trans. Inf. Forensics. Sec., 2016, 11, (11), pp. 2453–2465.
    14. 14)
      • 24. Cao, G., Iosifidis, A., Chen, K., et al: ‘Generalized multi-view embedding for visual recognition and cross-modal retrieval’, arXiv:1605.09696v1, 2016, pp. 1–13.
    15. 15)
      • 33. Iosifidis, A., Tefas, A., Pitas, I.: ‘Large-scale nonlinear facial image classification based on approximate kernel extreme learning machine’. IEEE Int. Conf. Image Processing, 2015.
    16. 16)
      • 4. Samaria, F., Harter, A.: ‘Parameterisation of a stochastic model for human face identification’. IEEE Workshop on Applications of Computer Vision, 1994.
    17. 17)
      • 22. Wong, W.K., Sun, M.: ‘Deep learning regularized fisher mappings’, IEEE Trans. Neural Netw., 2011, 22, (10), pp. 1668–1675.
    18. 18)
      • 31. Wolf, L., Hassner, T., Maoz, I.: ‘Face recognition in unconstrained videos with matched background similarity’. Computer Vision and Pattern Recognition, 2011.
    19. 19)
      • 27. Golub, G., Loan, C.: ‘Matrix computations’ (Johns Hopkins University Press, 1996, 3rd edn.).
    20. 20)
      • 20. Iosifidis, A., Gabbouj, M.: ‘Class-specific kernel discriminant analysis revisited: further analysis and extensions’, IEEE Trans. Cybern., 2016, DOI: 10.1109/TCYB.2016.2612479.
    21. 21)
      • 2. Li, Z., Park, U., Jain, A.: ‘A discriminative model for age invariant face recognition’, IEEE Trans. Inf. Forensics Sec., 2011, 6, (3), pp. 1028–1037.
    22. 22)
      • 8. Ye, J.: ‘Least squares linear discriminant analysis’. Int. Conf. Machine Learning, 2007.
    23. 23)
      • 34. Iosifidis, A., Tefas, A., Pitas, I.: ‘Approximate kernel extreme learning machine for large-scale data classification’, Neurocomputing, 2017, 219, pp. 210–220.
    24. 24)
      • 25. Jia, Y., Nie, F., Zhang, C.: ‘Trace ratio problem revisited’, IEEE Trans. Neural Netw., 2009, 20, (4), pp. 729–735.
    25. 25)
      • 6. Saul, L., Roweis, S.: ‘Think globally, fit locally: unsupervised learning of nonlinear manifolds’, J. Mach. Learn. Res., 2003, 4, pp. 119–155.
    26. 26)
      • 23. Stuhlsatz, A., Lippel, J., Zielke, T.: ‘Feature extraction with deep neural networks by a generalized discriminant analysis’, IEEE Trans. Neural Netw., 2012, 23, (4), pp. 596–608.
    27. 27)
      • 29. Li, Z., Lin, D., Tang, X.: ‘Nonparametric discriminant analysis for face recognition’, IEEE Trans. Pattern Anal. Mach. Intell., 2009, 31, (4), pp. 755–761.
    28. 28)
      • 16. Williams, C.K.I., Seeger, M.: ‘Using the Nyström method to speed up kernel machines’. Advances in Neural Information Processing Systems, 2001, pp. 682–688.
    29. 29)
      • 14. Arashloo, S., Kittler, J.: ‘Class-specific kernel fusion of multiple descriptors for face verification using multiscale binarized statistical image features’, IEEE Trans. Inf. Forensics Sec., 2014, 9, (12), pp. 2100–2109.
    30. 30)
      • 12. Li, Y.P., Kittler, J., Matas, J.: Face verification using client specific Fisher faces’, in Kent, J.T., Aykroyd, R.G. (Eds.): ‘Proceedings of International conference on The Statistics of Directions, Shapes and Images’ (University of Leeds, Dept. of Statistics, Leeds, UK, 2000), pp. 63–66.
    31. 31)
      • 30. Ortiza, E.G., Beckerb, B.C.: ‘Face recognition for web-scale datasets’, Comput. Vis. Image Underst., 2014, 118, pp. 153–170.
    32. 32)
      • 35. Rahimi, A., Recht, B.: ‘Random features for large-scale kernel machines’. Advances in Neural Information Processing Systems, 2007.
    33. 33)
      • 11. Iosifidis, A., Tefas, A., Pitas, I.: ‘Kernel reference discriminant analysis’, Pattern Recognit. Lett., 2014, 49, pp. 85–91.
    34. 34)
      • 15. Iosifidis, A., Tefas, A., Pitas, I.: ‘Class-specific reference discriminant analysis with application in human behavior analysis’, IEEE Trans. Human Mach. Syst., 2015, 45, (3), pp. 315–326.
    35. 35)
      • 32. Huang, G.B., Zhou, H., Ding, X., et al: ‘Extreme learning machine for regression and multi-class classification’, IEEE Trans. Syst. Man Cybern. B, 2012, 42, (2), pp. 513–529.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-bmt.2017.0081
Loading

Related content

content/journals/10.1049/iet-bmt.2017.0081
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading