Publication

Towards a Methodology for Training with Synthetic Data on the Example of Pedestrian Detection in a Frame-by-Frame Semantic Segmentation Task

Atanas Poibrenski, Janis Sprenger, Christian Müller

In: Proceedings of the 1st International Workshop on Software Engineering for AI in Autonomous Systems. International Workshop on Software Engineering for AI in Autonomous Systems (SEFAIAS-2018) located at ICSE May 27-June 3 Gothenburg Sweden Pages 31-34 SEFAIS '18 ISBN 978-1-4503-5739-5/18/05 ACM/IEEE 2018.

Abstract

In order to make highly/fully automated driving safe, synthetic training and validation data will be required, because critical road situations are too diverse and too rare. A few studies on using synthetic data have been published, reporting a general increase in accuracy. In this paper, we propose a novel method to gain more in-depth insights into the quality, performance, and influence of synthetic data during the training phase in a bounded setting. We demonstrate this method for the example of pedestrian detection in a frame-by-frame semantic segmentation class.

Projekte

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz