KIT | KIT-Bibliothek | Impressum | Datenschutz

Extending StructureNet to generate physically feasible 3D shapes

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

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 a nd 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. ... mehr

Open Access Logo

DOI: 10.5220/0010256702210228
Zugehörige Institution(en) am KIT Institut für Visualisierung und Datenanalyse (IVD)
Publikationstyp Proceedingsbeitrag
Publikationsjahr 2021
Sprache Englisch
Identifikator ISBN: 978-9897584886
KITopen-ID: 1000131333
Erschienen in Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 1: GRAPP. Ed.: A. A. Sousa
Veranstaltung 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2021), Online, 08.02.2021 – 10.02.2021
Verlag SciTePress
Seiten 221-228
Schlagwörter Generative Models, Shape Synthesis, Graph Neural Networks, Physical Constraints, Measure of Infeasibility
Nachgewiesen in Scopus
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page