pip install -r requirements.txt.- Download the training dataset from MS1M-ArcFace (85K ids/5.8M images).
- Download the verification dataset and labels from LFW.
To train a model, run /KD_Iresnet/train_kd.sh with the correct address.
To test a model, run mainOne.sh with the correct model path.
This project is made possible by the wonderful people and projects listed in this document.
- Fadi Boutros - Masked-Face-Recognition-KD
- Jia Guo - Insightface
- Jiankang Deng - Insightface
- Xiang An - Insightface
- Jack Yu - Insightface
- Baris Gecer - Insightface
@inproceedings{deng2019arcface,
title={Arcface: Additive angular margin loss for deep face recognition},
author={Deng, Jiankang and Guo, Jia and Xue, Niannan and Zafeiriou, Stefanos},
booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
pages={4690--4699},
year={2019}
}
@inproceedings{an2020partical_fc,
title={Partial FC: Training 10 Million Identities on a Single Machine},
author={An, Xiang and Zhu, Xuhan and Xiao, Yang and Wu, Lan and Zhang, Ming and Gao, Yuan and Qin, Bin and
Zhang, Debing and Fu Ying},
booktitle={Arxiv 2010.05222},
year={2020}
}
@INPROCEEDINGS{huber2021maskinvariant,
author={Huber, Marco and Boutros, Fadi and Kirchbuchner, Florian and Damer, Naser},
booktitle={2021 16th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021)},
title={Mask-invariant Face Recognition through Template-level Knowledge Distillation},
year={2021},
volume={},
number={},
pages={1-8},
doi={10.1109/FG52635.2021.9667081}
}