1.原文
![在这里插入图片描述](https://i-blog.csdnimg.cn/blog_migrate/cd802de6cbef2e916b589093b9c1cd1e.png)
2.步骤
3.踩坑过程
3.1 搭建环境
不能按官方文档写的直接pip,巨多坑
conda create -n face_recognition python=3.6
pip install -U opencv-python==3.4.4.19
pip install scipy==1.2.1 --user
face_recognition库需要安装dlib,pip不行,参考:
https://blog.csdn.net/HJZ11/article/details/107513672
https://www.cnblogs.com/ckAng/p/10981025.html
3.2 秒退、qt异常退出等
pip install 所有后若还不行:取消pyqt兼容性
![在这里插入图片描述](https://i-blog.csdnimg.cn/blog_migrate/b9e2b109c351c5babcf6a09b20c412e5.png)
3.3 AttributeError: module ‘scipy.misc’ has no attribute ‘imread’
pip install scipy==1.2.1 --user
3.4 TypeError: face_landmarks() got an unexpected keyword argument ‘model’
face_l = frc.face_landmarks(input)
3.5 opencv报错
pip install -U opencv-python==3.4.4.19