Low Resolution Face Recognition Using Lightweight VarGFaceNet Architecture with Adaptive Margin Loss

  • Daffa Tama Ramadani Universitas Singaperbangsa Karawang
  • Riza Ibnu Adam Universitas Singaperbangsa Karawang
  • Jajam Haerul Jaman Universitas Singaperbangsa Karawang
  • Chaerur Rozikin Universitas Singaperbangsa Karawang
  • G. Garno Universitas Singaperbangsa Karawang
Keywords: Face Recognition, Lightweight CNN, Low Resolution, Super Resolution

Abstract

Face recognition is a modern security solution that is quickly and easily integrated into most existing devices, so this system is widely applied to several domains as one of the security authorizations. Developing face recognition models using mainstream architectures (AlexNet, VGGNet, GoogleNet, ResNet, and SENet) will make it difficult to implement the models on mobile devices and embedded systems. In addition, low resolution images, such as those from CCTV surveillance cameras or drones, pose challenges for the models to recognize faces, as the images lack sufficient details for identification. Therefore, this research aims to analyze the performance of a face recognition model developed using the lightweight VarGFaceNet architecture with the adaptive margin loss AdaFace on a low-resolution image dataset. Based on the evaluation results on the LFW dataset, an accuracy of 99.08% was achieved on high-resolution data (112x112 pixels), while on the lowest synthetic low-resolution data (14x14 pixels), an accuracy of 79.87% was obtained with the assistance of the Real-ESRGAN and GFP-GAN super-resolution models. On the TinyFace dataset, without fine-tuning, a Rank-1 accuracy of 46.08% was achieved without using super-resolution models and 45.03% when utilizing super-resolution models.

Downloads

Download data is not yet available.

References

M. Heidari and K. Fouladi-Ghaleh, “Using Siamese Networks with Transfer Learning for Face Recognition on Small-Samples Datasets,” in 2020 International Conference on Machine Vision and Image Processing (MVIP), IEEE Computer Society, Feb. 2020. doi: 10.1109/MVIP49855.2020.9116915.

R. Prathivi and Y. Kurniawati, “Sistem Presensi Kelas Menggunakan Pengenalan Wajah Dengan Metode Haar Cascade Classifier,” Simetris: Jurnal Teknik Mesin, Elektro dan Ilmu Komputer, vol. 11, no. 1, pp. 135–142, Apr. 2020, doi: 10.24176/SIMET.V11I1.3754.

H. Muchtar and R. Apriadi, “Implementasi Pengenalan Wajah Pada Sistem Penguncian Rumah Dengan Metode Template Matching Menggunakan Open Source Computer Vision Library (Opencv),” RESISTOR (Elektronika Kendali Telekomunikasi Tenaga Listrik Komputer), vol. 2, no. 1, pp. 39–42, May 2019, doi: 10.24853/RESISTOR.2.1.39-42.

A. M. R, “Polri Kenalkan Sistem Pengawasan Buronan I-24/7 di Markas Interpol di Lyon,” detikNews, Dec. 04, 2022. https://news.detik.com/berita/d-6436076/polri-kenalkan-sistem-pengawasan-buronan-i-247-di-markas-interpol-di-lyon (accessed Apr. 12, 2023).

M. Yan, M. Zhao, Z. Xu, Q. Zhang, G. Wang, and Z. Su, “VarGFaceNet: An Efficient Variable Group Convolutional Neural Network for Lightweight Face Recognition,” in 2019 International Conference on Computer Vision Workshop, ICCVW 2019, Institute of Electrical and Electronics Engineers Inc., Oct. 2019, pp. 2647–2654. doi: 10.1109/ICCVW.2019.00323.

Q. Zhang et al., “VarGNet: Variable Group Convolutional Neural Network for Efficient Embedded Computing,” arXiv:1907.05653, Jul. 2019, [Online]. Available: http://arxiv.org/abs/1907.05653

Z. Cheng, X. Zhu, and S. Gong, “Low-Resolution Face Recognition,” arXiv:1811.08965v2, Nov. 2018, [Online]. Available: http://arxiv.org/abs/1811.08965

Y. Martínez-Díaz, H. Méndez-Vázquez, L. S. Luevano, L. Chang, and M. Gonzalez-Mendoza, “Lightweight low-resolution face recognition for surveillance applications,” in International Conference on Pattern Recognition, Milan: Institute of Electrical and Electronics Engineers Inc., Jan. 2020, pp. 5421–5428. doi: 10.1109/ICPR48806.2021.9412280.

M. Kim, A. K. Jain, and X. Liu, “AdaFace: Quality Adaptive Margin for Face Recognition,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), vol. 2022-June, pp. 18729–18738, Apr. 2022, doi: 10.1109/CVPR52688.2022.01819.

S. Chen, Y. Liu, X. Gao, and Z. Han, “MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices,” in Biometric Recognition, J. Zhou, Y. Wang, Z. Sun, Z. Jia, J. Feng, S. Shan, K. Ubul, and Z. Guo, Eds., Cham: Springer International Publishing, 2018, pp. 428–438.

Y. Martindez-DIaz, L. S. Luevano, H. Mendez-Vazquez, M. Nicolas-DIaz, L. Chang, and M. Gonzalez-Mendoza, “ShuffleFaceNet: A lightweight face architecture for efficient and highly-accurate face recognition,” in International Conference on Computer Vision Workshop, ICCVW 2019, Seoul: Institute of Electrical and Electronics Engineers Inc., Oct. 2019, pp. 2721–2728. doi: 10.1109/ICCVW.2019.00333.

M. Grgic, K. Delac, and S. Grgic, “SCface - Surveillance cameras face database,” Multimed Tools Appl, vol. 51, no. 3, pp. 863–879, Feb. 2011, doi: 10.1007/s11042-009-0417-2.

Z. Cheng, X. Zhu, and S. Gong, “Surveillance Face Recognition Challenge,” arXiv: 1804.09691, Apr. 2018, [Online]. Available: http://arxiv.org/abs/1804.09691

G. B. Huang, M. Ramesh, T. Berg, and E. Learned-Miller, “Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments,” Amherst, Oct. 2007. [Online]. Available: http://vis-www.cs.umass.edu/lfw/.

D. Yi, Z. Lei, S. Liao, and S. Z. Li, “Learning Face Representation from Scratch,” arXiv:1411.7923v1, Nov. 2014, [Online]. Available: http://arxiv.org/abs/1411.7923

A. Torralba, R. Fergus, and W. T. Freeman, “80 Million Tiny Images: A Large Data Set for Nonparametric Object and Scene Recognition,” IEEE Trans Pattern Anal Mach Intell, vol. 30, no. 11, pp. 1958–1970, 2008, doi: 10.1109/TPAMI.2008.128.

X. Wang, L. Xie, C. Dong, and Y. Shan, “Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data,” arXiv:2107.10833, pp. 1905–1914, Jul. 2021, doi: 10.1109/ICCVW54120.2021.00217.

X. Wang, Y. Li, H. Zhang, and Y. Shan, “Towards Real-World Blind Face Restoration with Generative Facial Prior,” Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 9164–9174, Jan. 2021, doi: 10.1109/CVPR46437.2021.00905.

Y. Martínez-Díaz et al., “Benchmarking lightweight face architectures on specific face recognition scenarios,” Artif Intell Rev, vol. 54, no. 8, pp. 6201–6244, 2021, doi: 10.1007/s10462-021-09974-2.

L. S. Luevano, L. Chang, H. Heydi Mendez-Vazquez, Y. Martinez-Diaz, and M. Gonzalez-Mendoza, “A Study on the Performance of Unconstrained Very Low Resolution Face Recognition: Analyzing Current Trends and New Research Directions,” IEEE Access, vol. 9, no. 9, pp. 75470–75493, May 2021, doi: 10.1109/ACCESS.2021.3080712.

InsightFace, “InsightFace: an open source 2D&3D deep face analysis library,” InsightFace. https://insightface.ai/ (accessed Apr. 27, 2023).

S. Z. Li and A. K. Jain, “Introduction,” in Handbook of Face Recognition, S. Z. Li and A. K. Jain, Eds., 2nd ed.London: Springer London, 2011, pp. 1–15. doi: 10.1007/978-0-85729-932-1.

D. E. Kurniawan, K. Adi, and A. F. Rohim, “Sistem Identifikasi Biometrika Wajah Menggunakan Metode Gabor KPCA dan Mahalanobis Distance,” JSINBIS J. Sist. Inf. Bisnis, vol. 2, no. 1, pp. 006–010, Jan. 2014, doi: 10.21456/vol2iss1pp006-010.

X. Wu, R. He, Z. Sun, and T. Tan, “A Light CNN for Deep Face Representation with Noisy Labels,” IEEE Transactions on Information Forensics and Security, vol. 13, no. 11, pp. 2884–2896, Nov. 2015, doi: 10.1109/TIFS.2018.2833032.

M. Wang and W. Deng, “Deep face recognition: A survey,” Neurocomputing, vol. 429, pp. 215–244, Mar. 2021, doi: 10.1016/J.NEUCOM.2020.10.081.

Microsoft Research, “DeepSpeed.” https://www.microsoft.com/en-us/research/project/deepspeed/ (accessed Jul. 31, 2023).

J. Deng, J. Guo, N. Xue, and S. Zafeiriou, “ArcFace: Additive Angular Margin Loss for Deep Face Recognition,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 4690–4699.

H. Wang et al., “CosFace: Large Margin Cosine Loss for Deep Face Recognition,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, Jan. 2018, pp. 5265–5274. doi: 10.1109/CVPR.2018.00552.

Published
2023-07-31
How to Cite
[1]
D. Ramadani, R. Adam, J. Jaman, C. Rozikin, and G. Garno, “Low Resolution Face Recognition Using Lightweight VarGFaceNet Architecture with Adaptive Margin Loss”, JAIC, vol. 7, no. 1, pp. 104-111, Jul. 2023.
Section
Articles

Most read articles by the same author(s)