MODERN APPROACHES TO FACE RECOGNITION IN LOW-LIGHT CONDITIONS: A REVIEW AND THE CONCEPT OF A HYBRID END-TO-END ARCHITECTURE
Abstract
The article addresses the problem of reliable face recognition in critical areas such as video surveillance and biometric authentication under low-light conditions. Existing approaches typically separate the tasks of image enhancement and face identification, which leads to error accumulation and loss of informative features. The aim of this work is to overcome this limitation by developing and theoretically substantiating a hybrid end-to-end architecture in which image enhancement and face recognition are solved jointly. The study provides a systematic review of modern methods, including classical algorithms (such as histogram equalization and noise suppression) and advanced deep neural networks (including EnlightenGAN, Zero-DCE, ArcFace, and RetinaFace). The main contribution is the integration of generative and identification modules into a single computational graph. The key result of the study is the demonstration that joint optimization of all processing stages within a unified model, unlike fragmented solutions, fundamentally changes the approach to the problem. Theoretical analysis and comparative evaluation of existing concepts show that the proposed architecture ensures a more efficient gradient flow during training, leading to the formation of higher-quality and noise-robust identity features. It is shown that this approach prevents error accumulation between stages and minimizes information loss. The novelty of the work lies in the holistic, end-to-end view of the face recognition problem under low-light conditions. The practical significance is confirmed by the applicability of the architecture in real systems, where its implementation can potentially improve reliability and processing speed by combining heterogeneous tasks into a single optimizable framework.
The article addresses the problem of reliable face recognition in critical areas such as video surveillance and biometric authentication under low-light conditions. Existing approaches typically separate the tasks of image enhancement and face identification, which leads to error accumulation and loss of informative features. The aim of this work is to overcome this limitation by developing and theoretically substantiating a hybrid end-to-end architecture in which image enhancement and face recognition are solved jointly. The study provides a systematic review of modern methods, including classical algorithms (such as histogram equalization and noise suppression) and advanced deep neural networks (including EnlightenGAN, Zero-DCE, ArcFace, and RetinaFace). The main contribution is the integration of generative and identification modules into a single computational graph. The key result of the study is the demonstration that joint optimization of all processing stages within a unified model, unlike fragmented solutions, fundamentally changes the approach to the problem. Theoretical analysis and comparative evaluation of existing concepts show that the proposed architecture ensures a more efficient gradient flow during training, leading to the formation of higher-quality and noise-robust identity features. It is shown that this approach prevents error accumulation between stages and minimizes information loss. The novelty of the work lies in the holistic, end-to-end view of the face recognition problem under low-light conditions. The practical significance is confirmed by the applicability of the architecture in real systems, where its implementation can potentially improve reliability and processing speed by combining heterogeneous tasks into a single optimizable framework.
##article.references##
1. Lopatin D.V. Issledovanie effektivnosti deskriptorov osobykh tochek na razlichnykh tipakh izobra-zheniy: magisterskaya dissertatsiya [Study of the Efficiency of Feature Point Descriptors on Various Types of Images: Master’s thesis]. Tol'yatti: TGU, 2022, 75 p.
2. Goncharov A.V., Karkishchenko A.N. Vliyanie osveshchennosti na kachestvo raspoznavaniya frontal'nykh lits [Influence of illumination on the quality of frontal face recognition], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2008, No. 4 (81), pp. 88-92.
3. Liang J., Li Y., Yang M., Xu Y. Recurrent Exposure Generation for Low-Light Face Detection, arXiv, 2020. arXiv:2007.10963. doi: 10.48550/arXiv.2007.10963.
4. Ying Z., Li G., Gao W. A New Low-Light Image Enhancement Algorithm Using Camera Response Model, Proceedings of the IEEE International Conference on Computer Vision Workshops (ICCV Workshops), 2017, pp. 3015-3023. doi: 10.1109/ICCVW.2017.354.
5. Wei C., Wang W., Yang W., Liu J. Deep Retinex Decomposition for Low-Light Enhancement (Retinex-Net), arXiv, 2018. arXiv:1808.04560. doi: 10.48550/arXiv.1808.04560.
6. Chen C., Chen Q., Xu J., Koltun V. Learning to See in the Dark (SID), Proceedings of the IEEE Confer-ence on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 3291-3300. doi: 10.1109/CVPR.2018.00347.
7. Jiang Y., Gong X., Liu D., Cheng Y., Fang C., Shen X., Yang J. Low-Light Image and Video Enhance-ment: A Comprehensive Survey, arXiv, 2022. arXiv:2212.10772. doi: 10.48550/arXiv.2212.10772.
8. Singh R., Kapoor K. Contrast Enhancement Techniques: A Survey, International Journal of Signal and Image Processing (IJSIP), 2017, Vol. 8, No. 3, pp. 45-56.
9. Kundu D., Chaudhuri B. Enhancement Measure Estimation (EME): A Metric for Image Contrast Evalu-ation, IET Image Processing, 2014, Vol. 8, No. 11, pp. 731-739. doi: 10.1049/iet-ipr.2013.0546.
10. Kovalevskiy V.A. Metody uluchsheniya izobrazheniy v sistemakh videonablyudeniya pri nedostatochnoy osveshchennosti: dissertatsiya [Methods for improving images in video surveillance systems under low illumination conditions: dissertation]. Moscow: MGTU im. Baumana, 2021, 112 p.
11. Guo C., Li C., Guo J., Loy C. C., Hou J., Kwong S., Cong R. Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement (Zero-DCE++), IEEE Transactions on Pattern Analysis and Ma-chine Intelligence, 2022, Vol. 44, No. 11, pp. 8825-8838. doi: 10.1109/TPAMI.2022.3141833.
12. Yang S., Luo P., Loy C. C., Tang X. WIDER FACE: A Face Detection Benchmark, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 5525-5533. doi: 10.1109/CVPR.2016.596.
13. Jain V., Learned-Miller E. FDDB: A Benchmark for Face Detection in Unconstrained Settings, Univer-sity of Massachusetts Amherst Technical Report, 2010, 32 p.
14. Maze B., Adams J., Duncan J. A., Kalka N., Grother P., Ngan M., Patterson E., Kao H.-C. IARPA Janus Benchmark–C: Face Dataset and Protocol, Proceedings of the International Conference on Bio-metrics (ICB), 2018, pp. 1-8. doi: 10.1109/ICB2018.2018.00012.
15. Patel O., Maravi Y.P.S., Sharma S. A Comparative Study of Histogram Equalization Based Image En-hancement Techniques for Brightness Preservation and Contrast Enhancement, Signal & Image Pro-cessing: An International Journal, 2013, Vol. 4, No. 1, pp. 1-10.
16. Bulat A., Tzimiropoulos G. How Far Are We from Solving the 2D & 3D Face Alignment Problem?, Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2488-2497. doi: 10.1109/ICCV.2017.270.
17. Yamashita R., Nishio M., Do R.K.G., Togashi K. Convolutional Neural Networks: An Overview and Application in Radiology, Insights into Imaging, 2018, Vol. 9, pp. 611-629. doi: 10.1007/s13244-018-0639-9.
18. Lore K.G., Akintayo A., Sarkar S. LLNet: A Deep Autoencoder Approach to Natural Low-Light Image Enhancement, Proceedings of the IEEE International Conference on Image Processing (ICIP), 2017, pp. 1422-1426. doi: 10.1109/ICIP.2017.8296512.
19. Jiang Y., Gong X., Liu D., Cheng Y., Fang C., Shen X., Yang J. EnlightenGAN: Deep Light Enhance-ment without Paired Supervision, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 2862-2871. doi: 10.1109/CVPR46437.2021.00473.
20. Deng J., Guo J., Zhou Y., Yu J., Kotsia I., Zafeiriou S. RetinaFace: Single-Shot Multi-Level Face Locali-sation in the Wild, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni-tion (CVPR), 2020, pp. 5203-5212. doi: 10.1109/CVPR42600.2020.00525.
21. Shen Y., Liu H., Qi G.-J., Song J., Tao D. FaceID-GAN: Learning a Symmetry Three-Player GAN for Identity-Preserving Face Synthesis, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 821-830. doi: 10.1109/CVPR.2018.00092.
22. Li C., Guo C., Loy C.C., Cong R. LightenFace: Realistic Face Relighting via GAN with Identity Loss, IEEE Transactions on Image Processing, 2022, Vol. 31, pp. 2551-2564. doi: 10.1109/TIP.2022.3157457.
23. Zamir S.W., Arora A., Khan S., Hayat M., Khan F.S., Yang M.H., Shao L. Learning Enriched Features for Real Image Restoration and Enhancement, European Conference on Computer Vision (ECCV), 2020, pp. 492-511. doi: 10.1007/978-3-030-58595-2_30.
24. Shi Y., Wang Z., Tang Z., Wang J., Yin Y. LighterFace Model for Community Face Detection and Recognition, IEEE Access, 2021, Vol. 9, pp. 50752-50762. doi: 10.1109/ACCESS.2021.3069343.
25. Singh P., Reibman A.R. Task-Aware Image Quality Estimators for Face Detection, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 6211-6220. doi: 10.1109/CVPR52688.2022.00614.
26. Liang X., Chen X., Ren K., Miao X., Chen Z., Jin Y. Low-Light Image Enhancement via Adaptive Fre-quency Decomposition Network, Scientific Reports, 2023, Vol. 13, Article No. 15208. doi: 10.1038/s41598-023-40899-8.
27. Zhang Y., Zhang J., Guo X. Kindling the Darkness: A Practical Low-Light Image Enhancer, Proceed-ings of the International Conference on Multimedia Modeling (MMM), 2021, pp. 133-145. doi: 10.1007/978-3-030-67832-6_12.
28. Xie Y., Ou J., Wen B., Yu Z., Tian W. A Joint Learning Method for Low-Light Facial Expression Recog-nition, Complex & Intelligent Systems, 2024, Vol. 10, pp. 2595-2611. doi: 10.1007/s40747-024-01762-z.
29. Grother P., Ngan M., Hanaoka K. Face Recognition Vendor Test (FRVT) Part 3: Demographic Effects, NIST Interagency Report 8280, 2019. Available at: https://nvlpubs.nist.gov/nistpubs/ ir/2019/NIST.IR.8280.pdf (accessed 23 April 2026).
30. George A., Ecabert C., Shahreza H.O., Kotwal K., Marcel S. EdgeFace: Efficient Face Recognition Model for Edge Devices, arXiv, 2023. arXiv:2307.01838. doi: 10.48550/arXiv.2307.01838.
