Extending StructureNet to Generate Physically Feasible 3D Shapes

Jannik Koch, Jannik Koch, Laura Haraké, Alisa Jung, Carsten Dachsbacher

2021

Abstract

StructureNet is a recently introduced n-ary graph network that generates 3D structures with awareness of geometric part relationships and promotes reasonable interactions between shape parts. However, depending on the inferred latent space, the generated objects may lack physical feasibility, since parts might be detached or not arranged in a load-bearing manner. We extend StructureNet’s training method to optimize the physical feasibility of these shapes by adapting its loss function to measure the structural intactness. Two new changes are hereby introduced and applied on disjunctive shape parts: First, for the physical feasibility of linked parts, forces acting between them are determined. Considering static equilibrium, compression and friction, they are assembled in a constraint system as the Measure of Infeasibility. The required interfaces between these parts are identified using Constructive Solid Geometry. Secondly, we define a novel metric called Hover Penalty that detects and penalizes unconnected shape parts to improve the overall feasibility. The extended StructureNet is trained on PartNet’s chair data set, using a bounding box representation for the geometry. We demonstrate first results that indicate a significant reduction of hovering shape parts and a promising correction of shapes that would be physically infeasible.

Paper Citation

in Harvard Style

Koch J., Haraké L., Jung A. and Dachsbacher C. (2021). Extending StructureNet to Generate Physically Feasible 3D Shapes. In Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 1: GRAPP; ISBN 978-989-758-488-6, SciTePress, pages 221-228. DOI: 10.5220/0010256702210228

in Bibtex Style

@conference{grapp21,
author={Jannik Koch and Laura Haraké and Alisa Jung and Carsten Dachsbacher},
title={Extending StructureNet to Generate Physically Feasible 3D Shapes},
booktitle={Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021) - Volume 1: GRAPP},
year={2021},
pages={221-228},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010256702210228},
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 1: GRAPP
TI - Extending StructureNet to Generate Physically Feasible 3D Shapes
SN - 978-989-758-488-6
AU - Koch J.
AU - Haraké L.
AU - Jung A.
AU - Dachsbacher C.
PY - 2021
SP - 221
EP - 228
DO - 10.5220/0010256702210228
PB - SciTePress