go back

Exploiting 3D Variational Autoencoders For Interactive Vehicle Design

Sneha Saha, Leandro Minku, Xin Yao, Bernhard Sendhoff, Stefan Menzel, "Exploiting 3D Variational Autoencoders For Interactive Vehicle Design ", 17th International Design Conference (Design 2022), 2022.

Abstract

In automotive digital development, 3D prototype creation is a team effort of designers and engineers, each contributing with creative ideas and technical design evaluations through means of computer simulations. To support the team in the 3D design ideation and exploration task, we propose an interactive 3D cooperative design system for assisted design explorations and faster performance estimations. We utilize the advantage of geometric deep learning-based variational autoencoders to create a low-dimensional latent manifold of 3D designs which is utilized within an interactive graphical user interface to guide 3D design generation and strengthen the decision-making process.



Download Bibtex file Download PDF

Search