TU Berlin

Communication Systems GroupScientific Publications

Page Content

to Navigation

Scientific Publications

Training a Convolutional Neural Network for Multi-Class Object Detection Using Solely Virtual World Data
Citation key 1491Bochinski2016
Author Erik Bochinski and Volker Eiselein and Thomas Sikora
Title of Book IEEE International Conference on Advanced Video and Signal-Based Surveillance
Pages 278–285
Year 2016
Address Colorado Springs, CO, USA
Month aug
Note Electronic ISBN: 978-1-5090-3811-4 Print on Demand(PoD) ISBN: 978-1-5090-3812-1 DOI: 10.1109/AVSS.2016.7738056
Abstract Convolutional neural networks are a popular choice for current object detection and classification systems. Their performance improves constantly but for effective training, large, hand-labeled datasets are required. We address the problem of obtaining customized, yet large enough datasets for CNN training by synthesizing them in a virtual world, thus eliminating the need for tedious human interaction for ground truth creation. We developed a CNN-based multi-class detection system that was trained solely on virtual world data and achieves competitive results compared to state-of-the-art detection systems.
Link to publication Download Bibtex entry


Quick Access

Schnellnavigation zur Seite über Nummerneingabe