To run the code, please first make sure you have all required packages installed. If you use conda, then just run (replace <env> with a proper environment name).
conda create --name <env> --file requirements.txt
or
This would set the name of the new environment as ml
conda env create -f ml.yml
After installed the environment, you can use conda activate <env> to activate the environment.
Then, remember to install PyQt5, you can use
conda install -c anaconda pyqt
You can also install all the required packages mannually by pip.
To run the application with GUI, use
python oweek_demo.py
To test the original model, use
python ./image_animation.py -i ./Inputs/feynman.jpeg -c ./checkpoints/vox-cpk.pth.tar
feynman.jpeg can be changed into other images in the Inputs folder.
To add more source images, just put new images into Inputs directory. The application will automatically detect them.
This application is based on face_recognition and Real_Time_Image_Animation