© The Institution of Engineering and Technology
A novel method to extract discriminative deep feature representations of facial images for face identification is presented. A new ‘multi-class pairwise discriminant loss’ is devised and incorporated it into the general deep convolutional neural network learning framework in a novel way, leading to highly discriminative deep face features. The method shows significant improvement over existing deep feature extraction techniques relying on softmax or triplet loss. Moreover, the method achieves a level of accuracy on the widely used identification protocols, which are better and comparable results than other state-of-the-art methods.
References
-
-
1)
-
3. Wu, X., He, R., Sun, Z.: ‘A lightened CNN for deep face representation’. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, June 2015.
-
2)
-
6. Chen, D., Cao, X., Wang, L., et al: ‘Bayesian face revisited: a joint formulation’. European Conf. Computer Vision (ECCV), Firenze, Italy, October 2012, pp. 566–579, .
-
3)
-
2. Schroff, F., Kalenichenko, D., Philbin, J.: ‘Facenet: a unified embedding for face recognition and clustering’. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, June 2015, pp. 815–823, .
-
4)
-
5. Wang, X., Tang, X.: ‘A unified framework for subspace face recognition’, Trans. Pattern Anal. Mach. Intell., 2014, 26, (9), pp. 1222–1228 (doi: 10.1109/TPAMI.2004.57).
-
5)
-
1. Parkhi, O.M., Vedaldi, A., Zisserman, A.: ‘Deep face recognition’. British Machine Vision Conf. (BMVC), Swansea, UK, September 2015, 1, (3), pp. 1–12, .
-
6)
-
7. Best-Rowden, L., Han, H., Otto, C., et al: ‘Unconstrained face recognition: identifying a person of interest from a media collection’, Trans. Inf. Forensics Sec., 2014, 9, (12), pp. 2144–2157 (doi: 10.1109/TIFS.2014.2359577).
-
7)
-
4. Taigman, Y., Yang, M., Ranzato, M., et al: ‘Web-scale training for face identification’. Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, June 2015, pp. 2746–2754, .
http://iet.metastore.ingenta.com/content/journals/10.1049/el.2017.2108
Related content
content/journals/10.1049/el.2017.2108
pub_keyword,iet_inspecKeyword,pub_concept
6
6