Share Email Print

Proceedings Paper

Optimized training of deep neural network for image analysis using synthetic objects and augmented reality
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Acquiring large amounts of data for training and testing Deep Learning (DL) models is time consuming and costly. The development of a process to generate synthetic objects and scenes using 3D graphics software is presented. By programming the path and environment in a 3D graphical engine, complex objects and scenes can be generated for the purpose of training and testing a Deep Neural Network (DNN) model in specific vision tasks. An automatic process has been developed to label and segment objects in synthetic images and generate their corresponding ground truth files. Performances of DNNs trained with synthetic data have been shown to outperform DNNs trained with real data.

Paper Details

Date Published: 13 May 2019
PDF: 11 pages
Proc. SPIE 10995, Pattern Recognition and Tracking XXX, 109950I (13 May 2019); doi: 10.1117/12.2522198
Show Author Affiliations
Thomas Lu, Jet Propulsion Lab. (United States)
Alexander Huyen, Jet Propulsion Lab. (United States)
Luan Nguyen, Jet Propulsion Lab. (United States)
Joseph Osborne, Jet Propulsion Lab. (United States)
Sarah Eldin, Jet Propulsion Lab. (United States)
Kyoungsik Yun, Jet Propulsion Lab. (United States)

Published in SPIE Proceedings Vol. 10995:
Pattern Recognition and Tracking XXX
Mohammad S. Alam, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?