Once you take a image of the face, the Android program will give the quality estimation value of it.
The SDD-FIQA face image quality estimation model was trained by Pytorch. You can get more information from: https://github.com/Tencent/TFace/tree/quality
The model was converted from .pth to .mnn, and it implements by MNN on Android, you can find some tips from: https://github.com/alibaba/MNN
The project was programmed for fun.