Radar-PointGNN: Graph Based Object Recognition for Unstructured Radar Point-cloud Data

Conference Paper (2021)
Author(s)

Peter Svenningsson (TU Delft - Microwave Sensing, Signals & Systems)

F. Fioranelli (TU Delft - Microwave Sensing, Signals & Systems)

O. Yarovyi (TU Delft - Microwave Sensing, Signals & Systems)

Microwave Sensing, Signals & Systems
Copyright
© 2021 P.O. Svenningsson, F. Fioranelli, Alexander Yarovoy
DOI related publication
https://doi.org/10.1109/RadarConf2147009.2021.9455172
More Info
expand_more
Publication Year
2021
Language
English
Copyright
© 2021 P.O. Svenningsson, F. Fioranelli, Alexander Yarovoy
Microwave Sensing, Signals & Systems
Bibliographical Note
Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.@en
ISBN (print)
978-1-7281-7610-9
ISBN (electronic)
978-1-7281-7609-3
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Perception systems for autonomous vehicles are reliant on a comprehensive sensor suite to identify objects in the environment. While object recognition systems in the LiDAR and camera modalities are reaching maturity, recognition models on sparse radar point measurements have remained an open research challenge. An object recognition model is here presented which imposes a graph structure on the radar point-cloud by connecting spatially proximal points and extracts local patterns by performing convolutional operations across the graph’s edges. The model’s performance is evaluated by the nuScenes benchmark and is the first radar object recognition model evaluated on the dataset. The results show that end-to-end deep learning solutions for object recognition in the radar domain are viable but currently not competitive with solutions based on LiDAR data.

Files

09455172.pdf
(pdf | 4.67 Mb)
- Embargo expired in 18-12-2021
License info not available