TopoFR: A Closer Look at Topology Alignment on Face Recognition

If you like TopoFR, please give us a star ⭐ on GitHub for the latest update~

This is the official PyTorch implementation of "[NeurIPS 2024] TopoFR: A Closer Look at Topology Alignment on Face Recognition".

Requirements

• Install Pytorch (torch>=1.9.0)
• pip install -r requirement.txt

Datasets

You can download the training datasets, including MS1MV2 and Glint360K:

• MS1MV2: Google Drive
• Glint360K: Baidu (code=:o3az)

You can download the test dataset IJB-C as follows:

• IJB-C: Google Drive

How to Train Models

1. You need to modify the path of training dataset in every configuration file in folder configs.

2. To run on a machine with 4 GPUs:

   python -m torch.distributed.launch --nproc_per_node=4 --nnodes=1 --node_rank=0 --master_addr="127.0.0.1" --master_port=12581 train.py 

How to Test Models

1. You need to modify the path of IJB-C dataset in eval_ijbc_ms1mv2.py and eval_ijbc_glint360k.py.

2. Run:

   python eval_ijbc_ms1mv2.py --model-prefix work_dirs/ms1mv2_r50/model.pt --result-dir work_dirs/ms1mv2_r50 --network r50 > ijbc_ms1mv2_R50_TopoFR.log 2>&1 &

   python eval_ijbc_glint360k.py --model-prefix work_dirs/glint360k_r50/model.pt --result-dir work_dirs/glint360k_r50 --network r50 > ijbc_glint360k_R50_TopoFR.log 2>&1 &

TopoFR Pretrained Models

• You can download the TopoFR models reported in our paper as follows:

• You can test the accuracy of these model: (e.g., Glint360K R100 TopoFR)

  python eval_ijbc_glint360k.py --model-prefix work_dirs/glint360k_r100/Glint360K_R100_TopoFR_9760.pt --result-dir work_dirs/glint360k_r100 --network r100 > ijbc_glint360k_R100_TopoFR.log 2>&1 &

Citation

• If you find it helpful for you, please consider citing our paper 📝 and giving a star ⭐.

  @article{dan2024topofr,
  title={TopoFR: A Closer Look at Topology Alignment on Face Recognition},
  author={Dan, Jun and Liu, Yang and Deng, Jiankang and Xie, Haoyu and Li, Siyuan and Sun, Baigui and Luo, Shan},
  journal={arXiv preprint arXiv:2410.10587},
  year={2024}
  }

Acknowledgments

We thank Insighface for the excellent code base.