当前位置：首页 > news >正文

学习insightface 的人脸识别

news 来源：原创 2025/6/9 7:23:49

1.Insightface_pytorch版本的自定义数据准备过程

https://github.com/artintel/LearningCode/blob/master/insightface_pytorch_datasets_make/README.md

https://github.com/artintel/LearningCode/tree/master/insightface_pytorch_datasets_make

2.mxnet_insightface开源代码使用——自定义数据集制作

https://zhuanlan.zhihu.com/p/92738758

3.insight face数据打包（一）：LFW

https://blog.csdn.net/ustczhng2012/article/details/113512500

4.Build-Your-Own-Face-Model

https://github.com/siriusdemon/Build-Your-Own-Face-Model/tree/master/recognition

https://github.com/siriusdemon/Build-Your-Own-Face-Model/blob/master/recognition/blog/data.md

5.arcface_torch

https://github.com/deepinsight/insightface/blob/master/recognition/arcface_torch/docs/prepare_custom_dataset.md

https://github.com/deepinsight/insightface/blob/master/recognition/arcface_torch/docs/prepare_custom_dataset.md

6.AI 实战 - 基于 insightface 的人脸识别

https://blog.csdn.net/qq_25439881/article/details/125179117

7.face recognition based on yolov8 & Arcface

https://github.com/JOKER-3-z/EASY_USE_face_recognition

8.yoloV5-arcface_forlearn

https://github.com/ooooxianyu/yoloV5-arcface_forlearn

9. yoloV5-arcface_forlearn-master

https://github.com/Lose-the-mind/yoloV5-arcface_forlearn-master

10.Pytorch-FaceNet-DogDataset

FaceNet Implementation on DogFace Dataset

https://github.com/fcernafukuzaki/Pytorch-FaceNet-DogDataset

Paper To Code implementation of Facenet in native pytorch on Dog Dataset

11.insightface pytorch 答疑

https://blog.csdn.net/minstyrain/article/details/126901471

快速上手 PyTorch 人脸相似度计算方法 ArcFace

基于insightface实现的人脸识别和人脸注册-CSDN博客

https://yeyupiaoling.blog.csdn.net/article/details/108304894

ArcFace详解与实践-CSDN博客

人脸识别从原理到实践

https://github.com/zhongyy/SFace/tree/main/SFace_torch

https://github.com/deepinsight/insightface/wiki/Dataset-Zoo

训练公开数据

先从insightface datasets下载数据，里面提供了对齐(112x112)好的人脸文件，并打包成rec格式加速读取，如果需要原始文件图片，可参考load_images_from_bin.py自行保存，也可使用image_iter_rec.py里的FaceDataset直接加载数据

insightface datasets:

https://github.com/deepinsight/insightface/tree/master/recognition/_datasets_

load_images_from_bin.py:

https://github.com/xuexingyu24/MobileFaceNet_Tutorial_Pytorch/blob/master/data_set/load_images_from_bin.py

image_iter_rec.py:

https://github.com/zhongyy/SFace/blob/main/SFace_torch/image_iter_rec.py

人脸识别0-10：insightFace-半监督系统搭建(5)-流程总结，代码分享。_insightface casia-facev5-CSDN博客

训练自己的数据

训练自己的数据要复杂很多，人半监督系统搭建(5)-流程总结首先通过爬虫等手段获取人脸图片，每个人放一个文件夹。https://github.com/DarkIsComing/1.FaceRecognition

MTCNN训练-CSDN博客

MTCNN训练

要部署的话可以使用MTCNN，

https://github.com/imistyrain/MTCNN

其提供了部署全平台实现，包括C++、python、ncnn和tensorflow，还有加速版本和opencv直接加载版本，是所有版本中的集大成者.

如果想了解算法原理，可以参考

MTCNN_Step_by_Step

https://github.com/xuexingyu24/MTCNN_Tutorial/blob/master/MTCNN_Step_by_Step.ipynb

本文的训练流程参考至

MTCNN_Training_Step_by_Step

https://github.com/xuexingyu24/MTCNN_Tutorial/blob/master/MTCNN_Training_Step_by_Step.ipynb

提供了主流的pytorch实现

训练数据准备

MTCNN混合了三类任务: 人脸/非人脸分类、人脸框回归和人脸关键点定位. WIDER FACE (左图) 用来进行人脸分类和框回归. CNN_FacePoint (右图) 用来进行关键点定位.

FaceNet Implementation on DogFace Dataset

相关文章：