A suitable conda environment named FPMT
can be created
and activated with:
conda env create -f environment.yaml
conda activate FPMT
The pre trained FPMT with L=3 model has been provided in folder weights
. Due to the parameter of FPMT being only 0.95M, its pth
file size is approximately 3.6M.
In the current directory, run the following command
python eval.py
The corresponding generated results have been placed in folder examples_results\test_pair
-
Face alignment. Please refer to the BeautyGAN or PSGAN code to crop the image according to the face landmarks.
-
Prepare face parsing. In our experiment, face parsing is generated by https://github.com/zllrunning/face-parsing.PyTorch.
-
Put face images in the
./examples/images/makeup
and./examples/images/non-makeup
. Put the results of face parsing in the./examples/seg1/makeup
and/examples/seg1/non-makeup
. -
Run
python eval.py
.