Repository files navigation

InsightFace project is mainly maintained ByJia Guo andJiankang Deng.

For all main contributors, please checkcontributing.

The code of InsightFace is released under the MIT License. There is no limitation for both academic and commercial usage.

The training data containing the annotation (and the models trained with these data) are available for non-commercial research purposes only.

Both manual-downloading models from our github repo and auto-downloading models with ourpython-library follow the above license policy(which is for non-commercial research purposes only).

2025-03-02inswapper-512-live launched.

2024-08-01 We’ve integrated our most advanced face-swapping models,inswapper_cyn andinswapper_dax, into thePicsi.Ai face-swapping service. These models outperform nearly all commercial alternatives and our previous open-source model (inswapper_128).

2024-05-04 We have addedInspireFace, which is a cross-platform face recognition SDK developed in C/C++, supporting multiple operating systems and various backends.

2022-08-12: We achieved Rank-1st ofPerspective Projection Based Monocular 3D Face Reconstruction ChallengeofECCV-2022 WCPA Workshop,paper andcode.

2021-10-29: We achieved 1st place on theVISA track ofNIST-FRVT 1:1 by using Partial FC (Xiang An, Jiankang Deng, Jia Guo).

2025-03-02inswapper-512-live launched.

2024-08-01 We’ve integrated our most advanced face-swapping models,inswapper_cyn andinswapper_dax, into thePicsi.Ai face-swapping service. These models outperform nearly all commercial alternatives and our previous open-source model (inswapper_128).

2024-05-04 We have addedInspireFace, which is a cross-platform face recognition SDK developed in C/C++, supporting multiple operating systems and various backends.

2024-04-17:Monocular Identity-Conditioned Facial Reflectance Reconstruction accepted byCVPR-2024.

2023-08-08: We released the implementation ofGeneralizing Gaze Estimation with Weak-Supervision from Synthetic Views atreconstruction/gaze.

2023-05-03: We have launched the ongoing version of wild face anti-spoofing challenge. See detailshere.

2023-04-01: We integrated our most advanced face-swapping models:inswapper_cyn andinswapper_dax and move the service to Discord bot, which also support editing on Midjourney generated images, see detail atweb-demos/swapping_discord and ourPicsi.Ai website.

2023-02-13: We launch a large scale in the wild face anti-spoofing challenge on CVPR23 Workshop, see details atchallenges/cvpr23-fas-wild.

2022-11-28: Single line code for facial identity swapping in our python packge ver 0.7, please check the examplehere.

2022-10-28:MFR-Ongoing website is refactored, please create issues if there's any bug.

2022-09-22: Now we haveweb-demos:face-localization,face-recognition, andface-swapping.

2022-08-12: We achieved Rank-1st ofPerspective Projection Based Monocular 3D Face Reconstruction ChallengeofECCV-2022 WCPA Workshop,paper andcode.

2022-03-30:Partial FC accepted by CVPR-2022.

2022-02-23:SCRFD accepted byICLR-2022.

2021-11-30:MFR-Ongoing challenge launched(same with IFRT), which is an extended version oficcv21-mfr.

2021-10-29: We achieved 1st place on theVISA track ofNIST-FRVT 1:1 by using Partial FC (Xiang An, Jiankang Deng, Jia Guo).

2021-10-11:Leaderboard ofICCV21 - Masked Face Recognition Challenge released. Video:Youtube,Bilibili.

2021-06-05: We launch aMasked Face Recognition Challenge & Workshop on ICCV 2021.

InsightFace is an open source 2D&3D deep face analysis toolbox, mainly based on PyTorch and MXNet.

Please check ourwebsite for detail.

The master branch works withPyTorch 1.6+ and/orMXNet=1.6-1.8, withPython 3.x.

InsightFace efficiently implements a rich variety of state of the art algorithms of face recognition, face detection and face alignment, which optimized for both training and deployment.

Please start with ourpython-package, for testing detection, recognition and alignment models on input images.

Please click the image to watch the Youtube video. For Bilibili users, clickhere.

Thepage on InsightFace website also describes all supported projects in InsightFace.

You may also interested in somechallenges hold by InsightFace.

In this module, we provide training data, network settings and loss designs for deep face recognition.

The supported methods are as follows:

• ArcFace_mxnet (CVPR'2019)
• ArcFace_torch (CVPR'2019)
• SubCenter ArcFace (ECCV'2020)
• PartialFC_mxnet (CVPR'2022)
• PartialFC_torch (CVPR'2022)
• VPL (CVPR'2021)
• Arcface_oneflow
• ArcFace_Paddle (CVPR'2019)

Commonly used network backbones are included in most of the methods, such as IResNet, MobilefaceNet, MobileNet, InceptionResNet_v2, DenseNet, etc..

The training data includes, but not limited to the cleaned MS1M, VGG2 and CASIA-Webface datasets, which were already packed in MXNet binary format. Pleasedataset page for detail.

We provide standard IJB and Megaface evaluation pipelines inevaluation

Please checkModel-Zoo for more pretrained models.

• TensorFlow:InsightFace_TF
• TensorFlow:tf-insightface
• TensorFlow:insightface
• PyTorch:InsightFace_Pytorch
• PyTorch:arcface-pytorch
• Caffe:arcface-caffe
• Caffe:CombinedMargin-caffe
• Tensorflow:InsightFace-tensorflow
• TensorRT:wang-xinyu/tensorrtx
• TensorRT:InsightFace-REST
• ONNXRuntime C++:ArcFace-ONNXRuntime
• ONNXRuntime Go:arcface-go
• MNN:ArcFace-MNN
• TNN:ArcFace-TNN
• NCNN:ArcFace-NCNN

In this module, we provide training data with annotation, network settings and loss designs for face detection training, evaluation and inference.

The supported methods are as follows:

• RetinaFace (CVPR'2020)
• SCRFD (Arxiv'2021)
• blazeface_paddle

RetinaFace is a practical single-stage face detector which is accepted byCVPR 2020. We provide training code, training dataset, pretrained models and evaluation scripts.

SCRFD is an efficient high accuracy face detection approach which is initialy described inArxiv. We provide an easy-to-use pipeline to train high efficiency face detectors with NAS supporting.

In this module, we provide datasets and training/inference pipelines for face alignment.

Supported methods:

• SDUNets (BMVC'2018)
• SimpleRegression

SDUNets is a heatmap based method which accepted onBMVC.

SimpleRegression provides very lightweight facial landmark models with fast coordinate regression. The input of these models is loose cropped face image while the output is the direct landmark coordinates.

If you findInsightFace useful in your research, please consider to cite the following related papers:

@inproceedings{ren2023pbidr,  title={Facial Geometric Detail Recovery via Implicit Representation},  author={Ren, Xingyu and Lattas, Alexandros and Gecer, Baris and Deng, Jiankang and Ma, Chao and Yang, Xiaokang},  booktitle={2023 IEEE 17th International Conference on Automatic Face and Gesture Recognition (FG)},    year={2023} }@article{guo2021sample,  title={Sample and Computation Redistribution for Efficient Face Detection},  author={Guo, Jia and Deng, Jiankang and Lattas, Alexandros and Zafeiriou, Stefanos},  journal={arXiv preprint arXiv:2105.04714},  year={2021}}@inproceedings{gecer2021ostec,  title={OSTeC: One-Shot Texture Completion},  author={Gecer, Baris and Deng, Jiankang and Zafeiriou, Stefanos},  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},  year={2021}}@inproceedings{an_2022_pfc_cvpr,  title={Killing Two Birds with One Stone: Efficient and Robust Training of Face Recognition CNNs by Partial FC},  author={An, Xiang and Deng, Jiangkang and Guo, Jia and Feng, Ziyong and Zhu, Xuhan and Jing, Yang and Tongliang, Liu},  booktitle={CVPR},  year={2022}}@inproceedings{an_2021_pfc_iccvw,  title={Partial FC: Training 10 Million Identities on a Single Machine},  author={An, Xiang and Zhu, Xuhan and Gao, Yuan and Xiao, Yang and Zhao, Yongle and Feng, Ziyong and Wu, Lan and Qin, Bin and Zhang, Ming and Zhang, Debing and Fu, Ying},  booktitle={ICCVW},  year={2021},}@inproceedings{deng2020subcenter,  title={Sub-center ArcFace: Boosting Face Recognition by Large-scale Noisy Web Faces},  author={Deng, Jiankang and Guo, Jia and Liu, Tongliang and Gong, Mingming and Zafeiriou, Stefanos},  booktitle={Proceedings of the IEEE Conference on European Conference on Computer Vision},  year={2020}}@inproceedings{Deng2020CVPR,title = {RetinaFace: Single-Shot Multi-Level Face Localisation in the Wild},author = {Deng, Jiankang and Guo, Jia and Ververas, Evangelos and Kotsia, Irene and Zafeiriou, Stefanos},booktitle = {CVPR},year = {2020}}@inproceedings{guo2018stacked,  title={Stacked Dense U-Nets with Dual Transformers for Robust Face Alignment},  author={Guo, Jia and Deng, Jiankang and Xue, Niannan and Zafeiriou, Stefanos},  booktitle={BMVC},  year={2018}}@article{deng2018menpo,  title={The Menpo benchmark for multi-pose 2D and 3D facial landmark localisation and tracking},  author={Deng, Jiankang and Roussos, Anastasios and Chrysos, Grigorios and Ververas, Evangelos and Kotsia, Irene and Shen, Jie and Zafeiriou, Stefanos},  journal={IJCV},  year={2018}}@inproceedings{deng2018arcface,title={ArcFace: Additive Angular Margin Loss for Deep Face Recognition},author={Deng, Jiankang and Guo, Jia and Niannan, Xue and Zafeiriou, Stefanos},booktitle={CVPR},year={2019}}

Main contributors:

• Jia Guo,guojia[at]gmail.com
• Jiankang Dengjiankangdeng[at]gmail.com
• Xiang Ananxiangsir[at]gmail.com
• Jack Yujackyu961127[at]gmail.com
• Baris Gecerbarisgecer[at]msn.com