This repository contains the official chumpy implementation of mocap body solver used for AMASS:
AMASS: Archive of Motion Capture as Surface Shapes
Naureen Mahmood, Nima Ghorbani, Nikolaus F. Troje, Gerard Pons-Moll, Michael J. Black
Full paper |
Video |
Project website |
Poster
This repository holds the code for MoSh++, introduced in AMASS, ICCV'19. MoSh++ is the upgraded version of MoSh, Sig.Asia'2014. Given a labeled marker-based motion capture (mocap) c3d file and the correspondences of the marker labels to the locations on the body, MoSh can return model parameters for every frame of the mocap sequence. The current MoSh++ code works with the following models:
The Current repository requires Python 3.7 and chumpy; a CPU based auto-differentiation package. This package is assumed to be used along with SOMA, the mocap auto-labeling package. Please install MoSh++ inside the conda environment of SOMA. Clone the moshpp repository, and run the following from the root directory:
sudo apt install libtbb-dev
pip install -r requirements.txt
cd src/moshpp/scan2mesh
sudo apt install libeigen3-dev
pip install -r requirements.txt
2. sudo apt install libtbb-dev
cd mesh_distance
make
cd ../../../..
python setup.py install
This repository is a complementary package to SOMA, an automatic mocap solver. Please refer to the SOMA repository for tutorials and use cases.
Please cite the following paper if you use this code directly or indirectly in your research/projects:
@inproceedings{AMASS:2019,
title={AMASS: Archive of Motion Capture as Surface Shapes},
author={Mahmood, Naureen and Ghorbani, Nima and Troje, Nikolaus F. and Pons-Moll, Gerard and Black, Michael J.},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
year={2019},
month = {Oct},
url = {https://amass.is.tue.mpg.de},
month_numeric = {10}
}
Please consider citing the initial version of MoSh from Loper et al. Sig. Asia'14:
@article{Loper:SIGASIA:2014,
title = {{MoSh}: Motion and Shape Capture from Sparse Markers},
author = {Loper, Matthew M. and Mahmood, Naureen and Black, Michael J.},
address = {New York, NY, USA},
publisher = {ACM},
month = nov,
number = {6},
volume = {33},
pages = {220:1--220:13},
journal = {ACM Transactions on Graphics, (Proc. SIGGRAPH Asia)},
url = {http://doi.acm.org/10.1145/2661229.2661273},
year = {2014},
doi = {10.1145/2661229.2661273}
}
Software Copyright License for non-commercial scientific research purposes. Please read carefully the terms and conditions and any accompanying documentation before you download and/or use the MoSh++ data and software, (the "Data & Software"), software, scripts, and animations. By downloading and/or using the Data & Software (including downloading, cloning, installing, and any other use of this repository), you acknowledge that you have read these terms and conditions, understand them, and agree to be bound by them. If you do not agree with these terms and conditions, you must not download and/or use the Data & Software. Any infringement of the terms of this agreement will automatically terminate your rights under this License.
The software is compiled using CGAL sources following the license in CGAL_LICENSE.pdf
The code in this repository is developed by Nima Ghorbani, Naureen Mahmood, and Matthew Loper while at Max-Planck Institute for Intelligent Systems, Tübingen, Germany.
If you have any questions you can contact us at [email protected].
For commercial licensing, contact [email protected]