KIT | KIT-Bibliothek | Impressum | Datenschutz

Strategies for supplementing recurrent neural network training for spatio-temporal prediction = Strategien zur Unterstützung des Trainings von Rekurrenten Neuronalen Netzen zur räumlich-zeitlichen Vorhersage

Schutera, Mark; Elser, Stefan; Abhau, Jochen; Mikut, Ralf; Reischl, Markus

Abstract (englisch):
In autonomous driving, prediction tasks address complex spatio-temporal data. This article describes the examination of Recurrent Neural Networks (RNNs) for object
trajectory prediction in the image space. The proposed methods enhance the performance and spatio-temporal prediction capabilities of Recurrent Neural Networks. Two different data augmentation strategies and a hyperparameter search are implemented for this purpose. A conventional data augmentation strategy and a Generative Adversarial Network (GAN) based strategy are analyzed with respect
to their ability to close the generalization gap of Recurrent Neural Networks. The results are then discussed using single-object tracklets provided by the KITTI Tracking
Dataset. This work demonstrates the benefits of augmenting spatio-temporal data with GANs.



Originalveröffentlichung
DOI: 10.1515/auto-2018-0124
Zugehörige Institution(en) am KIT Institut für Automation und angewandte Informatik (IAI)
Publikationstyp Zeitschriftenaufsatz
Jahr 2019
Sprache Englisch
Identifikator ISSN: 2196-677X, 0178-2312
KITopen-ID: 1000096468
HGF-Programm 47.01.02 (POF III, LK 01)
Erschienen in Automatisierungstechnik
Band 67
Heft 7
Seiten 545–556
Schlagworte Generative Adversarial Networks, data augmentation, Recurrent Neural Networks, generalization, trajectory prediction
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft
KITopen Landing Page