/ Machine Learning

A implementation of the 3D FLAME model in PyTorch

A implementation of the 3D FLAME model in PyTorch

FLAME

FLAME is a lightweight and expressive generic head model learned from over 33,000 of accurately aligned 3D scans. This repository provides sample PyTorch code to experiment with the FLAME model. For details how to adapt FLAME to images or 3D meshes, see the Chumpy or the Tensorflow FLAME repositories.

model_variations

FLAME combines a linear identity shape space (trained from 3800 scans of human heads) with an articulated neck, jaw, and eyeballs, pose-dependent corrective blendshapes, and additional global expression blendshapes. For details please about the model, please see the scientific publication and the supplementary video.

Installation

The code uses Python 3.7 and it is tested on PyTorch 1.4.

Setup FLAME PyTorch Virtual Environment

python3.7 -m venv <your_home_dir>/.virtualenvs/FLAME_PyTorch
source <your_home_dir>/.virtualenvs/FLAME_PyTorch/bin/activate

Clone the project and install requirements

git clone https://github.com/soubhiksanyal/FLAME_PyTorch
cd FLAME_PyTorch
pip install -r requirements.txt
mkdir model

Download models

  • Download FLAME model from here. You need to sign up and agree to the model license for access to the model. Copy the downloaded model inside the model folder.
  • Download Landmark embedings from RingNet Project. Copy it inside the model folder.

Demo

Loading FLAME and visualising the 3D landmarks on the face

Please note we used the pose dependent conture for the face as intriduced by RingNet Project.

Run the following command from the terminal

python main.py

License

FLAME is available under Creative Commons Attribution license. By using the model or the code code, you acknowledge that you have read the license terms (http://flame.is.tue.mpg.de/model_license), understand them, and agree to be bound by them. If you do not agree with these terms and conditions, you must not use the code.

Referencing FLAME

When using this code in a scientific publication, please cite

@article{FLAME:SiggraphAsia2017,
  title = {Learning a model of facial shape and expression from {4D} scans},
  author = {Li, Tianye and Bolkart, Timo and Black, Michael. J. and Li, Hao and Romero, Javier},
  journal = {ACM Transactions on Graphics, (Proc. SIGGRAPH Asia)},
  volume = {36},
  number = {6},
  year = {2017},
  url = {https://doi.org/10.1145/3130800.3130813}
}

Supported Projects

FLAME supports several projects such as

FLAME is part of SMPL-X: : A new joint 3D model of the human body, face and hands together

Contact

If you have any questions regarding the PyTorch implementation then you can contact us at [email protected] and [email protected].

Acknowledgement

This repository is build with modifications from SMPLX.

GitHub

Comments