3D Object Classification via Part Graphs

Florian Teich, Timo Lüddecke, Florentin Wörgötter

2021

Abstract

3D object classification often requires extraction of a global shape descriptor in order to predict the object class. In this work, we propose an alternative part-based approach. This involves automatically decomposing objects into semantic parts, creating part graphs and employing graph kernels on these graphs to classify objects based on the similarity of the part graphs. By employing this bottom-up approach, common substructures across objects from training and testing sets should be easily identifiable and may be used to compute similarities between objects. We compare our approach to state-of-the art methods relying on global shape description and obtain superior performance through the use of part graphs.

Download


Paper Citation


in Harvard Style

Teich F., Lüddecke T. and Wörgötter F. (2021). 3D Object Classification via Part Graphs. In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP; ISBN 978-989-758-488-6, SciTePress, pages 417-426. DOI: 10.5220/0010232604170426


in Bibtex Style

@conference{visapp21,
author={Florian Teich and Timo Lüddecke and Florentin Wörgötter},
title={3D Object Classification via Part Graphs},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP},
year={2021},
pages={417-426},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010232604170426},
isbn={978-989-758-488-6},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 5: VISAPP
TI - 3D Object Classification via Part Graphs
SN - 978-989-758-488-6
AU - Teich F.
AU - Lüddecke T.
AU - Wörgötter F.
PY - 2021
SP - 417
EP - 426
DO - 10.5220/0010232604170426
PB - SciTePress