Skip to content

Latest commit

 

History

History
29 lines (24 loc) · 1.04 KB

README.md

File metadata and controls

29 lines (24 loc) · 1.04 KB
Raw

FMA-3D

A method for adding a mask on a non-masked face image. Given a real masked face image (a) and a non-masked face image (d), we synthesize a photo-realistic masked face image with the mask from (a) and the facial area from (d). image

Some Results by FMA-3D

image

Requirements

  • python >= 3.7.1
  • pytorch >= 1.1.0

Usage

  • Extract the landmarks. You can extract the 106 landmarks by our face sdk or any other methods.
  • Add a mask on a face image. You can refer to add_mask_one.py as an example.
python add_mask_one.py

Note

If you use a different version of Python, you should rebuild the mesh_core_cython.so as descriped in face3d,

cd utils/mesh/cython
python setup.py build_ext -i

Reference

This project is mainly inspired by PRNet, face3d and pytorch-prnet