Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,7 @@
</p>

# News
- Our work [TransFace++](https://ieeexplore.ieee.org/document/11184862) got accepted to TPAMI ! (September 30th, 2025 UTC)
Comment on lines 12 to +13

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

This file contains two 'News' sections. This new item is being added to both, which is redundant. To improve readability and maintainability, it's recommended to have a single, consolidated 'News' section. Please consider removing this shorter, top-level 'News' section and keeping only the more comprehensive one that starts around line 40.

- Our work [FaceChain-MMID](https://www.sciencedirect.com/science/article/abs/pii/S0031320325005187) got accepted to Pattern Recognition ! (May 30th, 2025 UTC)
- More Technology Details of FaceChain-FACT train-free portrait generation can be seen in [Paper](https://arxiv.org/abs/2410.12312). (October 17th, 2024 UTC)
- Our work [TopoFR](https://arxiv.org/abs/2410.10587) got accepted to NeurIPS 2024 ! (September 26th, 2024 UTC)
Expand Down Expand Up @@ -37,6 +38,7 @@ FaceChain is powered by [ModelScope](https://github.com/modelscope/modelscope).


# News
- Our work [TransFace++](https://ieeexplore.ieee.org/document/11184862) got accepted to TPAMI ! (September 30th, 2025 UTC)
- Our work [FaceChain-MMID](https://www.sciencedirect.com/science/article/abs/pii/S0031320325005187) got accepted to Pattern Recognition ! (May 30th, 2025 UTC)
- More Technology Details of FaceChain-FACT train-free portrait generation can be seen in [Paper](https://arxiv.org/abs/2410.12312). (October 17th, 2024 UTC)
- Our work [TopoFR](https://arxiv.org/abs/2410.10587) got accepted to NeurIPS 2024 ! (September 26th, 2024 UTC)
Expand Down
2 changes: 2 additions & 0 deletions README_ZH.md
Original file line number Diff line number Diff line change
Expand Up @@ -6,6 +6,7 @@
<p>

# 最新消息
- Facechain算法创新工作[TransFace++](https://ieeexplore.ieee.org/abstract/document/11184862) 在期刊TPAMI上发表! (2025-09-30)
Comment on lines 8 to +9

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

This file contains two news sections, one under # 最新消息 and another under # News. This is redundant and the English heading in a Chinese README is inconsistent. To improve clarity, consider merging them into a single section under the Chinese heading (# 最新消息).

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The link for TransFace++ here (.../abstract/document/...) is different from the one in the English README.md (.../document/...). For consistency across documentation, it's recommended to use the same URL in both files.

Suggested change
- Facechain算法创新工作[TransFace++](https://ieeexplore.ieee.org/abstract/document/11184862) 在期刊TPAMI上发表! (2025-09-30)
- Facechain算法创新工作[TransFace++](https://ieeexplore.ieee.org/document/11184862) 在期刊TPAMI上发表! (2025-09-30)

- Facechain算法创新工作[FaceChain-MMID](https://www.sciencedirect.com/science/article/abs/pii/S0031320325005187) 在期刊Pattern Recognition上发表! (2025-05-30)
- 更多关于FaceChain-FACT免训练工作的技术细节可以在 [论文](https://arxiv.org/abs/2410.12312) 里查看. (2024-10-17)
- Facechain算法创新工作[TopoFR](https://arxiv.org/abs/2410.10587) 被NeurIPS 2024接收录用! (2024-09-26)
Expand All @@ -31,6 +32,7 @@ FaceChain的模型由[ModelScope](https://github.com/modelscope/modelscope)开


# News
- Facechain算法创新工作[TransFace++](https://ieeexplore.ieee.org/abstract/document/11184862) 在期刊TPAMI上发表! (2025-09-30)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The link for TransFace++ here (.../abstract/document/...) is different from the one in the English README.md (.../document/...). For consistency across documentation, it's recommended to use the same URL in both files.

Suggested change
- Facechain算法创新工作[TransFace++](https://ieeexplore.ieee.org/abstract/document/11184862) 在期刊TPAMI上发表! (2025-09-30)
- Facechain算法创新工作[TransFace++](https://ieeexplore.ieee.org/document/11184862) 在期刊TPAMI上发表! (2025-09-30)

- Facechain算法创新工作[FaceChain-MMID](https://www.sciencedirect.com/science/article/abs/pii/S0031320325005187) 在期刊Pattern Recognition上发表! (2025-05-30)
- 更多关于FaceChain-FACT免训练工作的技术细节可以在 [论文](https://arxiv.org/abs/2410.12312) 里查看. (2024-10-17)
- Facechain算法创新工作[TopoFR](https://arxiv.org/abs/2410.10587) 被NeurIPS 2024接收录用! (2024-09-26)
Expand Down
20 changes: 16 additions & 4 deletions face_module/TopoFR/README.md
Original file line number Diff line number Diff line change
@@ -1,8 +1,13 @@
<h2 align="center">TopoFR: A Closer Look at Topology Alignment on Face Recognition
<h5 align="center"> If you like TopoFR, please give us a star ⭐ on GitHub for the latest update~
<h5 align="center"> <a href='https://huggingface.co/spaces/developer0hye/TopoFR-Face-Recognition'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Demo-blue'></a>

</h2>
Comment on lines 1 to 5

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

high

The HTML structure here is invalid because heading elements like <h5> cannot be nested inside another heading element like <h2>. This can cause rendering issues on some markdown parsers. Please move the <h5> tags to be siblings of the <h2> tag, not children.


This is the official PyTorch implementation of ["[NeurIPS 2024] TopoFR: A Closer Look at Topology Alignment on Face Recognition"](https://arxiv.org/abs/2410.10587).
This is the official PyTorch implementation of ["[NeurIPS 2024] TopoFR: A Closer Look at Topology Alignment on Face Recognition"](https://proceedings.neurips.cc/paper_files/paper/2024/hash/419b6c974712adb884bfbbeea8e94d1b-Abstract-Conference.html).


[[Arxiv Version]](https://arxiv.org/abs/2410.10587) [[Code in Co-Author’s Rep.]](https://github.com/DanJun6737/TopoFR)[[机器之心]](https://www.jiqizhixin.com/articles/2024-10-17-6) [[CVer]](https://mp.weixin.qq.com/s/3NegnpJUGPfI_dOfkp_4LQ) [[CSDN]](https://blog.csdn.net/sunbaigui/article/details/143230746?spm=1001.2014.3001.5502)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

For better readability of the markdown source code, it's a good practice to add spaces between adjacent link definitions.

Suggested change
[[Arxiv Version]](https://arxiv.org/abs/2410.10587) [[Code in Co-Author’s Rep.]](https://github.com/DanJun6737/TopoFR)[[机器之心]](https://www.jiqizhixin.com/articles/2024-10-17-6) [[CVer]](https://mp.weixin.qq.com/s/3NegnpJUGPfI_dOfkp_4LQ) [[CSDN]](https://blog.csdn.net/sunbaigui/article/details/143230746?spm=1001.2014.3001.5502)
[[Arxiv Version]](https://arxiv.org/abs/2410.10587) [[Code in Co-Author’s Rep.]](https://github.com/DanJun6737/TopoFR) [[机器之心]](https://www.jiqizhixin.com/articles/2024-10-17-6) [[CVer]](https://mp.weixin.qq.com/s/3NegnpJUGPfI_dOfkp_4LQ) [[CSDN]](https://blog.csdn.net/sunbaigui/article/details/143230746?spm=1001.2014.3001.5502)


![image](docs/TopoFR.png)

Expand All @@ -26,6 +31,10 @@ You can download the test dataset IJB-C as follows:
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
```

## Hugging Face
You can easily experience our model on the Hugging Face space:
<a href='https://huggingface.co/spaces/developer0hye/TopoFR-Face-Recognition'><img src='https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Demo-blue'></a>

## 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.

Expand Down Expand Up @@ -66,12 +75,15 @@ python eval_ijbc_glint360k.py --model-prefix work_dirs/glint360k_r100/Glint360K_
* 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},
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},
journal={Advances in Neural Information Processing Systems},
volume={37},
pages={37213--37240},
year={2024}
}
```

## Acknowledgments
We thank Insighface for the excellent [code base](https://github.com/deepinsight/insightface/tree/master/recognition/arcface_torch).
We thank Insighface for the excellent [code base](https://github.com/deepinsight/insightface/tree/master/recognition/arcface_torch).
We would like to express our gratitude to the researcher [@developer0hye](https://github.com/developer0hye) for providing an outstanding [HF demo](https://huggingface.co/spaces/developer0hye/TopoFR-Face-Recognition).
2 changes: 1 addition & 1 deletion face_module/TransFace/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@

This is the official PyTorch implementation of [[ICCV-2023] TransFace: Calibrating Transformer Training for Face Recognition from a Data-Centric Perspective](https://openaccess.thecvf.com/content/ICCV2023/html/Dan_TransFace_Calibrating_Transformer_Training_for_Face_Recognition_from_a_Data-Centric_ICCV_2023_paper.html).

[[Arxiv Version]](https://arxiv.org/abs/2308.10133)
[[Arxiv Version]](https://arxiv.org/abs/2308.10133) [[Code in Co-Author's Rep.]](https://github.com/DanJun6737/TransFace) [[ModelScope]](https://www.modelscope.cn/models/iic/cv_vit_face-recognition) [[阿里云]](https://developer.aliyun.com/article/1319924) [[CVer]](https://mp.weixin.qq.com/s/n9HKqZ7DAt5Woo5BCwxq6w) [[CSDN]](https://blog.csdn.net/sunbaigui/article/details/136556006)



Expand Down