This repository contains the official implementation of the skeleton inference from:
OSSO: Obtaining Skeletal Shape from Outside
Marilyn Keller, Silvia Zuffi, Michael J. Black and Sergi Pujades
Full paper | Project website
News 15/02/2024: Check out our new skeleton model SKEL (Siggraph Asia 2023). SKEL as similar features as OSSO but the skin is rigged with the skeleton and the skeleton can be reposed.
Given a body shape with SMPL or STAR topology (in blue), we infer the underlying skeleton (in yellow).
Please follow the installation instruction in installation.md to setup all the required packages and models.
The skeleton can be inferred either from a SMPL or STAR mesh.
python main.py --mesh_input data/demo/body_female.ply --gender female -D -v
The final infered skeleton will be saved in the out folder and the intermediate meshes in out/tmp.
To export the skeleton mesh as separated meshes, add the --per_part option to the previous command.
If you find this model & software useful in your research, please consider citing:
@inproceedings{Keller:CVPR:2022,
title = {{OSSO}: Obtaining Skeletal Shape from Outside},
author = {Keller, Marilyn and Zuffi, Silvia and Black, Michael J. and Pujades, Sergi},
booktitle = {Proceedings IEEE/CVF Conf.~on Computer Vision and Pattern Recognition (CVPR)},
month = jun,
year = {2022},
month_numeric = {6}}
OSSO uses the Stitched Puppet by Silvia Zuffi and Michael J. Black, and the body model STAR by Ahmed Osman et al. The model was applied on AGORA (Priyanka Patel et al.) for demonstration.
This research has been conducted using the UK Biobank Resource under the Approved Project ID 51951. The authors thank the International Max Planck Research School for Intelligent Systems for supporting Marilyn Keller. Sergi Pujades’ work was funded by the ANR SEMBA project. We thank Anatoscope (www.anatoscope.com) for the initial skeleton mesh and useful discussions.
We also thank A. A. Osman for his helpful advice on body models, P. Patel for helping test OSSO on AGORA, T. McConnel, Y. Xiu, S. Tripathi and T. Yin for helping with the submission and release, and P. Ghosh, J. Tesch, A. Chandrasekaran, V. F. Abrevaya, S. Sanyal, O. Ben-Dov and P. Forte for fruitful discussions, advice and proofreading.
This code and model are available for non-commercial scientific research purposes as defined in the LICENSE.txt file.
For more questions, please contact osso@tue.mpg.de
For commercial licensing, please contact ps-licensing@tue.mpg.de