<<
| Conference/Proceedings | IEEE International Conference on Advanced Video and Signal-Based Surveillance |
| Start date | 23.08.2016 |
| End date | 26.08.2016 |
| Address | Colorado Springs, CO, USA |
| Pages | 278-285 |
| Author(s) | Erik Bochinski, Volker Eiselein, Thomas Sikora |
| Title | Training a Convolutional Neural Network for Multi-Class Object Detection Using Solely Virtual World Data |
| 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. |
| Key words | multimedia analysis, Deep Learning, Convolutional Neural Network, Object Detection, Dataset Synthesis, MOCATDataset |
| Note | Electronic ISBN: 978-1-5090-3811-4 Print on Demand(PoD) ISBN: 978-1-5090-3812-1 DOI: 10.1109/AVSS.2016.7738056 |
| File | 1491Bochinski2016.pdf |
BibTeX