CREATING A FACIAL RECOGNITION SYSTEM BASED ON A NEURAL NETWORK THAT WORKS IN REAL TIME

1
Lviv Polytechnic National University, Ukraine
2
Lviv Polytechnic National University
3
Lviv Polytechnic National University

The article is devoted to the theoretical foundations and practical aspects of facial recognition using neural networks. The paper analyzes various approaches to feature extraction from facial images, as well as methods for training neural networks for recognition. Special attention is paid to problems related to variations in lighting, facial expressions, and other factors affecting recog- nition accuracy. The paper describes the creation and optimization of a neural network for real-time human face recognition. Emphasis is placed on modern deep learning architectures, such as Convolutional Neural Networks (CNN), which provide high accuracy and processing speed. Key stages of the process are analyzed: data preprocessing, model training, use of hardware accelerators (e.g., GPU), and integration of algorithms into software for real-world applications. Special attention is given to challenges such as lighting variability, clothing changes, and data privacy protection. The article proposes effective methods for performance optimization and evaluates their impact on recognition accuracy. Experimental results demonstrate high processing speed, allowing the development of a model suitable for use in security systems, smartphones, and smart devices.

  1. PyTorch, 2024. [Online]. Available: https://pytorch.org/
  2. OpenCV, 2024. [Online]. Available: https://opencv.org/get- started/
  3. UML       Diagram, 2024.       [Online]. Available: https://app.diagrams.net/
  4. Material UI, 2024. [Online]. Available: https://mui.com/
  5. Pandas, 2024. [Online]. Available: https://pandas. pydata.org/.
  6. J. Deng, J. Guo, E. Ververas, I. Kotsia and S. Zafeiriou, "RetinaFace: Single-Shot Multi-Level Face Localization in the Wild", 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , Seattle, WA, USA, 2020, pp. 5202-5211.
  7. Xiangyu Zhu, Zhenhua Feng, Hailin Shi, Facial Landmark Localization , Handbook of Face Recognition, (137-169), 2023.
  8. V. Mrak, “Face recognition methods in video suveillange systems using machine learning”, Information and commu- nication technologies, electronic engineering, Lviv Polytech, Vol. 3, No. 2, pp. 33–42, 2023.
  9. Hang Du, Hailin Shi, Dan Zeng, Xiao-Ping Zhang, Tao Mei, “The Elements of End-to-end Deep Face Recognition: A Survey of Recent Advances”, ACM Computing Surveys, New York, USA, no. 54, 10s , p. 1-42, 2022.
  10. React, 2024. [Online]. Available: https://uk.reactjs.org/
  11. Typescript, 2024. [Online]. Available: https://www. typescriptlang.org/
  12. FastAPI, 2024. [Online]. Available: https://fastapi. tiangolo.com/
  13. Hayoung Jo, Seong-Whan Lee «Edge Conditional Node Update Graph Neural Network for Multi-variate Time Se- ries Anomaly Detection», Information Sciences no 679(12):121062. Doi:10.1016/j.ins.2024.121062.
  14. Kai Huang, Wei Gao, Real-time neural network inference on extremely weak devices: agile offloading with explainable AI, MobiCom '22: Proceedings of the 28th Annual Interna- tional Conference on Mobile Computing And Networking. Pages 200 – 213. https://doi.org/10.1145/3495243.3560551