Identity-Aware Lightweight MobileNetV2 with Distillation and Optuna for Face Spoofing Detection
DOI:
https://doi.org/10.29408/edumatic.v10i2.34863Keywords:
face anti-spoofing, hyperparameter optimization, knowledge distillation, mobilenetv2, optunaAbstract
Presentation attacks, commonly known as face spoofing, remain a major security challenge in facial recognition-based authentication systems because forged media such as printed photos and replayed videos can deceive biometric verification. Lightweight CNN models such as MobileNetV2 are suitable for practical implementation, but their limited representational capacity may affect their ability to capture subtle spoofing cues and generalize to unseen identities. Previous evaluations may also produce inflated performance estimates when images from the same identity appear across training and testing sets. This study evaluates Knowledge Distillation and Optuna-based hyperparameter tuning on MobileNetV2 for lightweight face anti-spoofing under an identity-aware evaluation protocol. The novelty lies in an identity-aware comparison between representation enhancement through EfficientNet-B0-based Knowledge Distillation and optimization-based improvement through Optuna. A total of 60,000 CelebA-Spoof images were divided using an 80:10:10 subject-disjoint split, and four scenarios were compared. The baseline MobileNetV2 achieved the best overall balance, with an accuracy of 0.9943, F1-score of 0.9958, and ACER of 0.0058. Meanwhile, Knowledge Distillation obtained the lowest APCER of 0.0035, indicating fewer spoof samples were incorrectly accepted as live under the identity-aware evaluation setting.
References
Cai, R., Yu, Z., Kong, C., Li, H., Chen, C., Hu, Y., & Kot, A. C. (2024). S-Adapter: Generalizing Vision Transformer for Face Anti-Spoofing With Statistical Tokens. IEEE Transactions on Information Forensics and Security, 19, 8385–8397. https://doi.org/10.1109/TIFS.2024.3420699
Gong, L. Y., Li, X. J., & Chong, P. H. J. (2024). Facial Anti-Spoofing Using “Clue Maps.” Sensors, 24(23). https://doi.org/10.3390/s24237635
Gulzar, Y. (2023). Fruit Image Classification Model Based on MobileNetV2 with Deep Transfer Learning Technique. Sustainability (Switzerland), 15(3). https://doi.org/10.3390/su15031906
Guo, Z., Wang, D., He, Q., & Zhang, P. (2024). Leveraging logit uncertainty for better knowledge distillation. Scientific Reports, 14(1). https://doi.org/10.1038/s41598-024-82647-6
Ham, H., Saptawijaya, A., & Arymurthy, A. M. (2025). Domain Generalization and Domain Adaptation Approaches for Face Anti-Spoofing: A Survey. IEEE Access, 13, 149390–149408. https://doi.org/10.1109/ACCESS.2025.3592034
Kong, Z., Zhang, W., Wang, T., Zhang, K., Li, Y., Tang, X., & Luo, W. (2024). Dual Teacher Knowledge Distillation with Domain Alignment for Face Anti-Spoofing. IEEE Transactions on Circuits and Systems for Video Technology, 34(12), 13177–13189. https://doi.org/10.1109/TCSVT.2024.3451294
Lai, L. H., Lin, Y. L., Liu, Y. H., Lai, J. P., Yang, W. C., Hou, H. P., & Pai, P. F. (2024). The Use of Machine Learning Models with Optuna in Disease Prediction. Electronics (Switzerland), 13(23). https://doi.org/10.3390/electronics13234775
Li, Y., Sun, W., Li, Z., & Guo, X. (2025). Face Anti-Spoofing Based on Adaptive Channel Enhancement and Intra-Class Constraint. Journal of Imaging, 11(4). https://doi.org/10.3390/jimaging11040116
Li, Z., Cai, R., Li, H., Lam, K. Y., Hu, Y., & Kot, A. C. (2022). One-Class Knowledge Distillation for Face Presentation Attack Detection. IEEE Transactions on Information Forensics and Security, 17, 2137–2150. https://doi.org/10.1109/TIFS.2022.3178240
Mun, H. J., & Lee, M. H. (2022). Design for Visitor Authentication Based on Face Recognition Technology Using CCTV. IEEE Access, 10, 124604–124618. https://doi.org/10.1109/ACCESS.2022.3223374
Qi, H., Han, R., Duan, K., Shi, Y., Qi, X., Gao, C., & Ren, L. (2025). A high-performance adaptive fusion network for face anti-spoofing detection. Scientific Reports, 15(1). https://doi.org/10.1038/s41598-025-21461-0
Shahreza, H. O., George, A., & Marcel, S. (2024). Knowledge Distillation for Face Recognition Using Synthetic Data With Dynamic Latent Sampling. IEEE Access, 12, 187800–187812. https://doi.org/10.1109/ACCESS.2024.3505621
Wang, W., Liu, P., Zheng, H., Ying, R., & Wen, F. (2023). Domain Generalization for Face Anti-Spoofing via Negative Data Augmentation. IEEE Transactions on Information Forensics and Security, 18, 2333–2344. https://doi.org/10.1109/TIFS.2023.3266138
Xiao, J., Wang, W., Zhang, L., & Liu, H. (2024). A MobileFaceNet-Based Face Anti-Spoofing Algorithm for Low-Quality Images. Electronics (Switzerland), 13(14). https://doi.org/10.3390/electronics13142801
Xing, H., Tan, S. Y., Qamar, F., & Jiao, Y. (2025). Face Anti-Spoofing Based on Deep Learning: A Comprehensive Survey. In Applied Sciences (Switzerland) (Vol. 15, Number 12). Multidisciplinary Digital Publishing Institute (MDPI). https://doi.org/10.3390/app15126891
Xu, M., Yoon, S., Fuentes, A., & Park, D. S. (2023). A Comprehensive Survey of Image Augmentation Techniques for Deep Learning. Pattern Recognition, 137(137), 109347. https://doi.org/10.1016/j.patcog.2023.109347
Yu, Z., Qin, Y., Li, X., Zhao, C., Lei, Z., & Zhao, G. (2023). Deep Learning for Face Anti-Spoofing: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(5), 5609–5631. https://doi.org/10.1109/TPAMI.2022.3215850
Zawar, R., & Chakkarwar, V. (2023). Real-Time Face Liveness Detection and Face Anti-spoofing Using Deep Learning. International Journal of Intelligent Systems and Applications in Engineering, 11, 626–636. https://doi.org/10.2991/978-94-6463-196-8_47
Zeng, G. (2025). Invariance Properties and Evaluation Metrics Derived from the Confusion Matrix in Multiclass Classification. Mathematics, 13(16). https://doi.org/10.3390/math13162609
Zheng, T., Li, B., Wu, S., Wan, B., Mu, G., Liu, S., Ding, S., & Wang, J. (2024). MFAE: Masked Frequency Autoencoders for Domain Generalization Face Anti-Spoofing. IEEE Transactions on Information Forensics and Security, 19, 4058–4069. https://doi.org/10.1109/TIFS.2024.3371266
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Alif Sahputra, Alva Hendi Muhammad
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All articles in this journal are the sole responsibility of the authors. Edumatic: Jurnal Pendidikan Informatika can be accessed free of charge, in accordance with the Creative Commons license used.
This work is licensed under a Lisensi a Creative Commons Attribution-ShareAlike 4.0 International License.
