A Face Recognition Pipeline Based on MTCNN for Detection and ResNet for Feature Extraction
Keywords:
MTCNN algorithm, ResNet algorithm, Face testing, Face recognition, Migration LearningAbstract
With the ongoing development of artificial intelligence and the continuous advancement of machine learning technologies, face recognition has become widely adopted in intelligent identification systems. In many campus environments, however, attendance systems continue to rely on traditional manual appointment and check-in methods, which incur significant time costs and negatively affect classroom efficiency. To address this issue, this study employs the MTCNN (Multi-Task Cascaded Convolutional Network) algorithm for face detection and the ResNet algorithm for face recognition, applying transfer learning to the classroom attendance task. Experimental comparisons were conducted with GoogLeNet, AlexNet, and VGG. The results demonstrate that ResNet achieves the highest face recognition accuracy among the models evaluated, indicating its strong potential for deployment in attendance systems.
References
Pinyoanuntapong, Ekkasit, et al. "Gaitsada: Self-aligned domain adaptation for mmwave gait recognition." 2023 IEEE 20th International Conference on Mobile Ad Hoc and Smart Systems (MASS). IEEE, 2023.
Narouei, F. H., Tang, Z., Wang, S. I., Hashmi, R. H., Welch, D., Sethuraman, S., ... & McNeill, V. F. (2025). Effects of germicidal far-UVC on ozone and particulate matter in a conference room. Plos one, 20(8), e0328224.
Xiao, J., Hai, L., Li, Y., Li, H., Gong, M., Wang, Z., ... & He, D. (2022). An Ultrasmall Fe3O4‐Decorated Polydopamine Hybrid Nanozyme Enables Continuous Conversion of Oxygen into Toxic Hydroxyl Radical via GSH‐Depleted Cascade Redox Reactions for Intensive Wound Disinfection. Small, 18(9), 2105465.
Zhao, X., Zhang, L., & Hu, Z. (2023). Smart warehouse track identification based on Res2Net-YOLACT+HSV. Innovation & Technology Advances, 1(1), 7–11. https://doi.org/10.61187/ita.v1i1.2
Deng, X., Yang, Y., Wang, B., Zhang, X., Huang, S., Zhang, Y., & Lu, Q. (2026). LLM-MVR: LLM-Guided Multi-View Reasoning Distillation for Sarcasm Detection. Available at SSRN 6795979.
Wang, Y., Jiang, K., Zhang, T., Tian, K., & Jiang, G. (2026). QA-ReID: Quality-Aware Query-Adaptive Convolution Leveraging Fused Global and Structural Cues for Clothes-Changing ReID. arXiv preprint arXiv:2601.19133.
Shan, X., Xu, Y., Xia, T., & Lin, Y. S. (2025, October). Rethinking Wine Tasting for Chinese Consumers: A Service Design Approach Enhanced by Multimodal Personalization. In 2025 International Conference on Content-Based Multimedia Indexing (CBMI) (pp. 1-5). IEEE.
Xia, T., Xu, Y., & Shan, X. (2025, May). KOA-Monitor: A Digital Intervention and Functional Assessment System for Knee Osteoarthritis Patients. In International Conference on Human-Computer Interaction (pp. 388-405). Cham: Springer Nature Switzerland.
Ge, H., & Wu, Y. (2023). An Empirical Study of Adoption of ChatGPT for Bug Fixing among Professional Developers. Innovation & Technology Advances, 1(1), 21–29. https://doi.org/10.61187/ita.v1i1.19
Wu, Y., Liang, P., Xiang, Y., Yuan, M., Liu, J., Yang, J., ... & Yan, W. (2026, March). Tiny-Critic RAG: Empowering Agentic Fallback with Parameter-Efficient Small Language Models. In 2026 9th International Conference on Advanced Algorithms and Control Engineering (ICAACE) (pp. 2577-2580). IEEE.
Yan, W., Wu, Y., Liang, P., Yuan, M., Liu, J., Yang, J., & Li, X. (2026, March). PRISM: Pipeline for Root-cause Investigation via Specialized Multi-agents. In 2026 International Conference on Generative Artificial Intelligence and Information Security (GAIIS) (pp. 709-712). IEEE.
Yuan, M., Liu, J., Yang, J., Li, X., Yan, W., Wu, Y., & Liang, P. (2026, March). TA-Mem: Tool-Augmented Autonomous Memory Retrieval for LLM in Long-Term Conversational QA. In 2026 9th International Conference on Advanced Algorithms and Control Engineering (ICAACE) (pp. 2684-2688). IEEE.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Xiangyu Shen
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
