/3DDFA

Primary LanguagePythonMIT LicenseMIT

3DDFA

This is a PyTorch reimplementation of the paper: Face Alignment in Full Pose Range: A 3D Total Solution.

Dataset

MS-Celeb-1M dataset for training, 3,804,846 faces over 85,164 identities.

Dependencies

  • Python 3.6.8
  • PyTorch 1.3.0

Usage

  1. Clone this repo.
  2. Build cython module:
$ cd utils/cython
$ python3 setup.py build_ext -i

Data preprocess

Extract images, scan them, to get bounding boxes and landmarks:

$ python3 extract.py
$ python3 pre_process.py

Train

$ python3 train.py

To visualize the training process：

$ tensorboard --logdir=runs

Demo

$ python3 demo.py

Image

image

3D Face

image