3D Object Classification and Parameter Estimation based on Parametric Procedural Models

Roman Getto, Kenten Fina, Lennart Jarms, Arjan Kuijper, Dieter Fellner

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

Abstract

Classifying and gathering additional information about an unknown 3D objects is dependent on having a large amount of learning data. We propose to use procedural models as data foundation for this task. In our method we (semi-)automatically define parameters for a procedural model constructed with a modeling tool. Then we use the procedural models to classify an object and also automatically estimate the best parameters. We use a standard convolutional neural network and three different object similarity measures to estimate the best parameters at each degree of detail. We evaluate all steps of our approach using several procedural models and show that we can achieve high classification accuracy and meaningful parameters for unknown objects.
Original languageEnglish
Title of host publicationWSCG 2018
EditorsVaclav Skala
PublisherUniversity of West Bohemia
ISBN (Print)978-80-86943-40-4
Publication statusPublished - 2018
Event26th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision - Pilsen, Czech Republic
Duration: 28 May 20181 Jun 2018

Publication series

NameComputer Science Research Notes
Volume2801

Conference

Conference26th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision
Abbreviated titleWSCG 2018
CountryCzech Republic
CityPilsen
Period28/05/181/06/18

Keywords

  • Procedural modeling
  • Parametric modeling
  • Parameterization
  • 3D Objects
  • Classifications
  • Deep learning
  • Guiding Theme: Digitized Work
  • Research Area: Computer graphics (CG)

Fields of Expertise

  • Information, Communication & Computing

Cite this

Getto, R., Fina, K., Jarms, L., Kuijper, A., & Fellner, D. (2018). 3D Object Classification and Parameter Estimation based on Parametric Procedural Models. In V. Skala (Ed.), WSCG 2018 (Computer Science Research Notes ; Vol. 2801). University of West Bohemia.