Exploring the application of neural networks for facial image reconstruction in recognition systems
https://doi.org/10.15514/ISPRAS-2022-34(6)-8
Abstract
Identifying a person in a digital image using computer vision is a crucial aspect of this field. The presence of external objects, such as medical masks that cover part of the face, can drastically reduce recognition accuracy and increase errors from 5% to 50%, depending on the algorithm. This paper investigates the use of neural networks, in particular the generative adversarial network (GAN), to solve the problem of reconstructing an image of a face covered by a medical mask to improve face recognition accuracy.
About the Authors
Evgeny Igorevich MARKIN
Penza State Technological University
Russian Federation
Russian Federation
Candidate of Technical Sciences, Assistant of the Department of Programming
Valentina Vladimirovna ZUPAROVA
Penza State Technological University
Russian Federation
Russian Federation
Postgraduate Student of the Department of Programming
Alexey Ivanovich MARTYSHKIN
Penza State Technological University
Russian Federation
Russian Federation
Candidate of Technical Sciences, Associate Professor, Head of the Department of Programming
References
1. Goldstein A.J., Harmon L.D., Lesk A.B. Identification of human faces. Proceedings of the IEEE, vol. 59, issue 5, 1971, pp. 748–760.
2. Ngan M.L., Grother P.J., Hanaoka K.K. Ongoing Face Recognition Vendor Test (FRVT) Part 6B: Face recognition accuracy with face masks using post-COVID-19 algorithms. NIST Interagency/Internal Report (NISTIR) no. 8331, 2020, 83 p.
3. Блейхут Р.Э.Быстрые алгоритмы цифровой обработки сигналов. М., Мир, 1989 г., 448 cтр. / Blahut R.E. Fast Algorithms for Digital Signal Processing. Addison-Wesley, 1985, 455 p.
4. Оппенгейм А., Шафер Р. Цифровая обработка сигналов. М., Техносфера, 2019 г., 1048 cтр. / Oppenheim A.V., Schafer R.W. Discrete-Time Signal Processing. Pearson, 2009, 1144 p.
5. Бершадская Е.Г., Маркин Е.И., Мартышкин А.И. Методы идентификации личности по изображению лица. XXI век итоги прошлого и проблемы настоящего плюс, том 9, вып. 1, стр. 49-53 / Bershadskaya E.G., Markin E.I., Martyshkin A.I. Methods for personal image identification. XXI Century: Resumes of the Past and Challenges of the Present plus, vol. 9, issue 1, pp. 49-53 (in Russian).
6. Choi J., Han B. MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators. arXiv2107.07260 , 2021, 20 p.
7. Adler J., Lunz S. Banach wasserstein GAN. In Proc. of the 32nd International Conference on Neural Information Processing Systems, 2018, pp. 6755-6764.
8. Arjovsky M., Bottou L. Towards principled methods for training generative adversarial networks. arXiv1701.04862, 2017, 17 p.
9. Маркин Е.И., Мартышкин А.И., Зупарова В.В. Анализ возможностей нейронных сетей для генерации фотореалистичных изображений. Современные информационные технологии, вып. 33, 2021, стр. 30-34 / Маркин Е.И., Мартышкин А.И., Zuparova V.V. Analysis of the capabilities of neural networks to generate photorealistic images. Modern Information Technology, vol. 33, 2021, pp. 30-34 (in Russian).
10. Chi N.N.K. Active shape models their training and application. Bachelor’ Thesis. International University HCMC, Vietnam, 2011.
11. Gulrajani I., Ahmed F. et al. Improved training of wasserstein GANs. In Proc. of the 31st International Conference on Neural Information Processing Systems, 2017, pp. 5769-5779.
12. Salimans T., Goodfellow I. et al. Improved techniques for training GANs. In Proc. of the 30st International Conference on Neural Information Processing Systems, 2016, pp. 2234-2242.
13. Huszár F. How (not) to train your generative model: Scheduled sampling, likelihood, adversary? arXiv1511.05101, 2015, 9p.
14. Zhang W., Liu Y. et al. RankSRGAN: Super Resolution Generative Adversarial Networks with Learning to Rank. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 10, 2022, pp. 7149-7166.
15. Zhou E., Fan H. et al. Extensive facial landmark localization with coarse-to-fine convolutional network cascade. In Proc. of the IEEE International Conference on Computer Vision Workshops, 2013, pp. 386-391.
16. Wei J., Liu M. et al. DuelGAN: A Duel Between Two Discriminators Stabilizes the GAN Training. Lecture Notes in Computer Science, vol. 13683, 2022, pp. 290-317.
17. Wang Z., Bovik A.C. et al. Image quality assessment: from error visibility to structural similarity. IEEE Transactions on Image Processing, vol. 13, issue 4, 2004, pp. 600-612.
18. Yamashita R., Nishio M. et al. Convolutional neural networks: an overview and application in radiology. Insights into Imaging, vol. 9, issue 4, 2018, pp. 611-629.
19. Heo J.Three-dimensional generic elastic models for two-dimensional pose synthesis and face recognition Proquest, Umi Dissertation Publishing, 2011, 154 p.
Review
For citations:
MARKIN E.I., ZUPAROVA V.V., MARTYSHKIN A.I. Exploring the application of neural networks for facial image reconstruction in recognition systems. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2022;34(6):117-126. (In Russ.) https://doi.org/10.15514/ISPRAS-2022-34(6)-8
Email this article 