3D Car Shape Reconstruction from a Single Sketch Image

Nozawa, Naoiki, Shum, Hubert, Ho, Edmond and Morishima, Shigeo (2019) 3D Car Shape Reconstruction from a Single Sketch Image. In: Proceedings - MIG 2019: ACM Conference on Motion, Interaction, and Games: Newcastle upon Tyne, England, October 28-30, 2019. Association for Computing Machinery, New York, p. 37. ISBN 9781450369947

Preview

Text Nozawa et al - 3D Car Shape Reconstruction from a Single Sketch Image AAM.pdf - Accepted Version Download (1MB) | Preview

Abstract

Efficient car shape design is a challenging problem in both the automotive industry and the computer animation/games industry. In this paper, we present a system to reconstruct the 3D car shape from a single 2D sketch image. To learn the correlation between 2D sketches and 3D cars, we propose a Variational Autoencoder deep neural network that takes a 2D sketch and generates a set of multiview depth & mask images, which are more effective representation comparing to 3D mesh, and can be combined to form the 3D car shape. To ensure the volume and diversity of the training data, we propose a feature-preserving car mesh augmentation pipeline for data augmentation. Since deep learning has limited capacity to reconstruct fine-detail features, we propose a lazy learning approach that constructs a small subspace based on a few relevant car samples in the database. Due to the small size of such a subspace, fine details can be represented effectively with a small number of parameters. With a low-cost optimization process, a high-quality car with detailed features is created. Experimental results show that the system performs consistently to create highly realistic cars of substantially different shape and topology, with a very low computational cost.

Actions (login required)

View Item

Downloads