Face Recognition Based on Improved FaceNet Model
The convolutional neural networks (CNN) is one of the most successful deep learning model in the field of face recognition, the different image regions are always treated equally when extracting image features, but in fact different parts of the face play different roles in face recognition. For overcoming this defect, a weighted average pooling algorithm is proposed in this paper, the different weights are assigned to the abstract features from different local image regions in the pooling operation, so as to reflect its different roles in face recognition. The weighted average pooling algorithm is applied to the FaceNet network, and a face recognition algorithm based on the improved FaceNet model is proposed. The simulation experiments show that the proposed face recognition algorithm has higher recognition accuracy than the existing face recognition methods based on deep learning.
KeywordsFace recognition Deep learning FaceNet Convolutional neural networks Weighting coefficient
This work was supported by the Shaanxi Natural Science Foundation (2016JQ5051) and the Department of Education Shaanxi Province (2013JK1023).
- 1.Mao, Y.: Research on Face Recognition Algorithm Based on Deep Neural Network. Master, Zhejiang University (2017)Google Scholar
- 2.Jing, C., Song, T., Zhuang, L., Liu, G., Wang, L., Liu, K.: A survey of face recognition technology based on deep convolutional neural networks. Comput. Appl. Softw. 35(1), 223–231 (2018). https://doi.org/10.3969/j.issn.1000-386x.2018.01.039CrossRefGoogle Scholar
- 5.Qin, H., Yan, J., Li, X., Hu, X.: Joint training of cascaded CNN for face detection. In: 29th IEEE Conference on Computer Vision and Pattern Recognition, pp. 3456–3465. IEEE, Las Vegas (2016). https://doi.org/10.1109/cvpr.2016.376
- 7.Taigman, Y., Yang, M., Ranzato, M., Wolf, L.: DeepFace: Closing the gap to human-level performance in face verification. In: 27th IEEE Conference on Computer Vision and Pattern Recognition, pp. 1701–1708. IEEE, Columbus (2014). https://doi.org/10.1109/CVPR.2014.220
- 8.Sun, Y., Wang, X., Tang, X.: Deep learning face representation from predicting 10,000 classes. In: 27th IEEE Conference on Computer Vision and Pattern Recognition, pp. 1891–1898. IEEE, Columbus (2014). https://doi.org/10.1109/cvpr.2014.244
- 9.Sun, Y., Chen, Y., Wang, X., Tang, X.: Deep learning face representation by joint identification-verification. In: 28th Annual Conference on Neural Information Processing Systems 2014, pp. 1988–1996. Neural information processing systems foundation, Montreal (2014)Google Scholar
- 10.Sun, Y., Wang, X., Tang, X.: Deeply learned face representations are sparse, selective, and robust. In: 28th IEEE Conference on Computer Vision and Pattern Recognition, pp. 2892–2900. IEEE, Boston (2015). https://doi.org/10.1109/cvpr.2015.7298907
- 11.Sun, Y., Ding, L., Wang, X., Tang, X.: DeepID3: Face recognition with very deep neural networks. arXiv:1502.00873 (2015)
- 12.Schroff, F., Kalenichenko, D., Philbin, J.: FaceNet: a unified embedding for face recognition and clustering. In: 28th IEEE Conference on Computer Vision and Pattern Recognition, pp. 815–823. IEEE, Boston (2015). https://doi.org/10.1109/cvpr.2015.7298682
- 14.Yi, D., Lei, Z., Liao, S., Li, S. Z.: Learning face representation from scratch. arXiv:1411.7923 (2014)
- 15.Huang, G.B., Ramesh, M., Berg, T., Learned-Miller, E.: Labeled faces in the wild: a database for studying face recognition in unconstrained environments. Technical report, University of Massachusetts, Amherst (2007)Google Scholar