Prior-Less 3D Human Shape Reconstruction with an Earth Mover's Distance Informed CNN

Jingtian Zhang, Hubert P. H. Shum, Kevin D. McCay and Edmond S. L. Ho
Proceedings of the 2019 ACM SIGGRAPH Conference on Motion, Interaction and Games (MIG) Posters, 2019

Prior-Less 3D Human Shape Reconstruction with an Earth Mover's Distance Informed CNN

Abstract

We propose a novel end-to-end deep learning framework, capable of 3D human shape reconstruction from a 2D image without the need of a 3D prior parametric model. We employ a "prior-less" representation of the human shape using unordered point clouds. Due to the lack of prior information, comparing the generated and ground truth point clouds to evaluate the reconstruction error is challenging. We solve this problem by proposing an Earth Mover’s Distance (EMD) function to find the optimal mapping between point clouds. Our experimental results show that we are able to obtain a visually accurate estimation of the 3D human shape from a single 2D image, with some inaccuracy for heavily occluded parts.


Downloads


YouTube


Cite This Research

Plain Text

Jingtian Zhang, Hubert P. H. Shum, Kevin D. McCay and Edmond S. L. Ho, "Prior-Less 3D Human Shape Reconstruction with an Earth Mover's Distance Informed CNN," in MIG '19: Proceedings of the 2019 ACM SIGGRAPH Conference on Motion, Interaction and Games, pp. 44:1-44:2, Newcastle upon Tyne, UK, ACM, Oct 2019.

BibTeX

@inproceedings{zhang19priorless,
 author={Zhang, Jingtian and Shum, Hubert P. H. and McCay, Kevin D. and Ho, Edmond S. L.},
 booktitle={Proceedings of the 2019 ACM SIGGRAPH Conference on Motion, Interaction and Games},
 series={MIG '19},
 title={Prior-Less 3D Human Shape Reconstruction with an Earth Mover's Distance Informed CNN},
 year={2019},
 month={10},
 pages={44:1--44:2},
 numpages={2},
 doi={10.1145/3359566.3364694},
 isbn={978-1-4503-6994-7},
 publisher={ACM},
 Address={New York, NY, USA},
 location={Newcastle upon Tyne, UK},
}

RIS

TY  - CONF
AU  - Zhang, Jingtian
AU  - Shum, Hubert P. H.
AU  - McCay, Kevin D.
AU  - Ho, Edmond S. L.
T2  - Proceedings of the 2019 ACM SIGGRAPH Conference on Motion, Interaction and Games
TI  - Prior-Less 3D Human Shape Reconstruction with an Earth Mover's Distance Informed CNN
PY  - 2019
Y1  - 10 2019
SP  - 44:1
EP  - 44:2
DO  - 10.1145/3359566.3364694
SN  - 978-1-4503-6994-7
PB  - ACM
ER  - 


Supporting Grants

Northumbria University

Postgraduate Research Scholarship (Ref: ): £65,000, Principal Investigator ()
Received from Faculty of Engineering and Environment%s, Northumbria University, UK, 2015-2018
Project Page

Similar Research

Ruizhi Liu, Paolo Remagnino and Hubert P. H. Shum, "Neural-Code PIFu: High-Fidelity Single Image 3D Human Reconstruction via Neural Code Integration", Proceedings of the 2024 International Conference on Pattern Recognition (ICPR), 2024
Shanfeng Hu, Hubert P. H. Shum and Antonio Mucherino, "DSPP: Deep Shape and Pose Priors of Humans", Proceedings of the 2019 ACM SIGGRAPH Conference on Motion, Interaction and Games (MIG), 2019
Ziyi Chang, George Alex Koulieris and Hubert P. H. Shum, "3D Reconstruction of Sculptures from Single Images via Unsupervised Domain Adaptation on Implicit Models", Proceedings of the 2022 ACM Symposium on Virtual Reality Software and Technology (VRST), 2022
Naoki Nozawa, Hubert P. H. Shum, Edmond S. L. Ho and Shigeo Morishima, "3D Car Shape Reconstruction from a Single Sketch Image", Proceedings of the 2019 ACM SIGGRAPH Conference on Motion, Interaction and Games (MIG) Posters, 2019

HomeGoogle ScholarYouTubeLinkedInTwitter/XGitHubORCIDResearchGateEmail
 
Print