Quantifying The Generative Capabilities Of Variational Autoencoders For 3D Car Point Clouds

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Authors

  • Sneha Saha
  • Stefan Menzel
  • Bernhard Sendhoff
  • Patricia Wollstadt

Colleges, School and Institutes

External organisations

  • Honda Research Institute Europe GmbH, Germany
  • Southern University of Science and Technology

Abstract

During each cycle of automotive development, large
amounts of geometric data are generated as results of design
studies and simulation tasks. Discovering hidden knowledge from
this data and making it available to the development team
strengthens the design process by utilizing historic information
when creating novel products. To this end, we propose to
use powerful geometric deep learning models that learn lowdimensional
representation of the design data in an unsupervised
fashion. Trained models allow to efficiently explore the design
space, as well as to generate novel designs. One popular class
of generative models are variational autoencoders, which have
however been rarely applied to geometric data. Hence, we use
a variational autoencoder for 3D point clouds (PC-VAE) and
explore the model’s generative capabilities with a focus on the
generation of realistic yet novel 3D shapes. We apply the PC-VAE
to point clouds sampled from car shapes from a benchmark
data set and employ quantitative measures to show that our
PC-VAE generates realistic car shapes, wile returning a richer
variety of unseen shapes compared to a baseline autoencoder.
Finally, we demonstrate how the PC-VAE can be guided towards
generating shapes with desired target properties by optimizing
the parameters that maximize the output of a trained classifier
for said target properties. We conclude that generative models
are a powerful tool that may aid designers in automotive product
development.

Details

Original languageEnglish
Title of host publicationIEEE Symposium Series on Computational Intelligence (SSCI 2020) - Proceedings
Publication statusAccepted/In press - 18 Sep 2020
Event2020 IEEE Symposium Series on Computational Intelligence (SSCI 2020) - Canberra, Australia
Duration: 1 Dec 20204 Dec 2020

Conference

Conference2020 IEEE Symposium Series on Computational Intelligence (SSCI 2020)
Country/TerritoryAustralia
CityCanberra
Period1/12/204/12/20

Keywords

  • Representation learning, geometric deep learning, point clouds, generative model, novelty