Perceiving Systems Mesh Package
This package contains core functions for manipulating meshes and visualizing them.
It requires Python 3.5+ and is supported on Linux and macOS operating systems.
The Mesh processing libraries support several of our projects such as
- CoMA: Convolutional Mesh Encoders for Generating 3D Faces
- FLAME: Learning a model of facial shape and expression from 4D scans
- MANO: Modeling and Capturing Hands and Bodies Together
- SMPL: A Skinned Multi-Person Linear Model
- VOCA: Voice Operated Character Animation
- RingNet: 3D Face Shape and Expression Reconstruction from an Image
Requirements
You first need to install the Boost <http://www.boost.org>_ libraries.
You can compile your own local version or simply do on Linux:
$ sudo apt-get install libboost-dev
On macOS:
$ brew install boost
Installation
First, create a dedicated Python virtual environment and activate it:
$ python3 -m venv --copies my_venv
$ source my_venv/bin/activate
You should then compile and install the psbody-mesh package using the Makefile.
If you are using the system-wide Boost libraries:
$ make all
or the libraries locally installed:
$ BOOST_ROOT=/path/to/boost/libraries make all
Testing
To run the tests, simply do:
$ make tests
Documentation
A detailed documentation can be compiled using the Makefile:
$ make documentation
License
Please refer for LICENSE.txt for using this software. The software is compiled using CGAL sources following the license in CGAL_LICENSE.pdf
Acknowledgments
We thank the external contribution from the following people: