In today’s world, face recognition is very much important in modern biometric and surveillance applications,as well
as its accuracy.Accuracy gets reduced because of various challenges in real-world like changes in illumination, facial ex-
pressions and also because of presence of multiple faces in a single image frame. This work explains an approach which is
based on Genetic Algorithm (GA) to improve multiple face recognition; this can be done by the optimization of both feature
selection and classifier parameters. Initially Face Net is used to extract face embeddings and then GA is applied to them to
identify the most informative subsets of features. A chromosome is used to represent every candidate solution which is
evaluated with a multi-objective fitness function that considers accuracy, computational efficiency and robustness with pose
and occlusion variations.The implemented framework is scalable and can be used in various applications like authentication,
surveillance and access control.
[1] M. Turk and A. Pentland, “Eigenfaces for recognition,” Journal of Cognitive Neuroscience, vol. 3, no. 1, pp. 71–86,
1991.
[2] T. Ahonen, A. Hadid, and M. Pietika¨inen, “Face description using local binary patterns and its application to
recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 28, no. 12, pp. 2037–2041, 2006.
[3] P. N. Belhumeur, J. P. Hespanha, and D. J. Kriegman, “Eigenfaces versus Fisherfaces: Class-specific linear projection
for recognition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 711–720, 1997.
[4] G. B. Huang, M. Ramesh, T. Berg, and E. Learned-Miller, “Labelled Faces in the Wild: Database for unconstrained
face recognition,” University of Massachusetts, Amherst, 2007.
[5] I. Jolliffe, Principal Component Analysis, Springer, 2002.
[6] J. H. Holland, Adaptation in Natural and Artificial Systems, MIT Press, 1992.
[7] D. E. Goldberg, Genetic Algorithms in Search, Optimization, and Machine Learning, Addison-Wesley, 1989.
[8] K. A. De Jong, “A ten year perspective on genetic algorithms,” in Proc. Int. Conf. on Genetic Algorithms, 1985, pp.
169–177.
[9] H. Almuallim and T. G. Dietterich, “Learning Boolean concepts amid irrelevant features,” Artificial Intelligence, vol.
69, no. 1–2, pp. 279–305, 1994.
[10] Y. Freund and R. Schapire, “A decision-theoretic generalization of online learning and boosting,” Journal of
Computer and System Sciences, vol. 55, no. 1, pp. 119–139, 1997.
