In this work, we developed a lightweight face recognition model for low-resolution face images with VarGFaceNet architecture and adaptive margin loss AdaFace. This work also use super resolution model such as Real-ESRGAN and GFP-GAN to eliminate degradation in low-resolution images and reconstruct poses and facial expressions. for details please check, Low Resolution Face Recognition Using Lightweight VarGFaceNet Architecture with Adaptive Margin Loss.
If this work help you, please cite.
@article{Ramadani_Adam_Jaman_Rozikin_Garno_2023,
title={Low Resolution Face Recognition Using Lightweight VarGFaceNet Architecture with Adaptive Margin Loss},
volume={7},
url={https://jurnal.polibatam.ac.id/index.php/JAIC/article/view/5831},
DOI={10.30871/jaic.v7i1.5831},
number={1},
journal={Journal of Applied Informatics and Computing},
author={Ramadani, Daffa and Adam, Riza and Jaman, Jajam and Rozikin, Chaerur and Garno, G.},
year={2023},
month={Jul.},
pages={104-111}}
- Original VarGFaceNet: https://github.com/zma-c-137/VarGFaceNet
- VarGFaceNet-Pytorch: https://github.com/enquan/VarGFaceNet_Pytorch
- Base Code: https://github.com/mbfaria/ShuffleFaceNet_Pytorch
- AdaFace: https://github.com/mk-minchul/AdaFace
- Real-ESRGAN: https://github.com/xinntao/Real-ESRGAN
- GFP-GAN: https://github.com/TencentARC/GFPGAN