This repo holds the pytorch improved version of the paper: Learning-Robust-3D-Face-Reconstruction-and-Discriminative-Identity-Representation
Requirements
. Pytorch >=1.0
. Python >= 3.6
. Platform: Linux
- Clone this repo
git clone https://github.com/Cluoyao/Learning-Robust-3D-Face-Reconstruction-and-Discriminative-Identity-Representation.git - Training dataset
Firstly, download train and test dataset train_aug_120x120.zip Baidu Yun or Google Drive and AFLW-2000-3D dataset Bidu Yun or Google Drive
Then, download pre-trained model in Baidu Yun Keywords:pymr, and put it into training_debug/logs
After that, you can run the verify_recognition.py to test face recognition.