Hyperspectral 3D Point Cloud Segmentation Using RandLA-Net
Autor(en): | Mitschke, Isaak Wiemann, Thomas Igelbrink, Felix Hertzberg, Joachim |
Herausgeber: | Petrovic, I. Markovic, I. Menegatti, E. |
Stichwörter: | Deep learning; Hyperspectral imaging; Point clouds; Semantic segmentation | Erscheinungsdatum: | 2023 | Herausgeber: | Springer Science and Business Media Deutschland GmbH | Enthalten in: | Lecture Notes in Networks and Systems | Band: | 577 LNNS | Startseite: | 301 – 312 | Zusammenfassung: | Point clouds are commonly used in robotics to represent 3D maps. To gain further understanding of their content, it is useful to annotate such maps semantically. To segment 3D point clouds with RGB values, different solutions exist. In machine learning, pre-trained classifiers are used for this purpose. Since it is not always possible to differentiate between entities relying solely on RGB information, hyperspectral histograms can augment the 3D data. The aim of this work is to evaluate, if hyperspectral information can improve the segmentation results for ambiguous objects, e.g., streets, sidewalks, and cars using established deep learning methods. Given the reported performance on geometrical 3D data and the possibility to directly integrate point annotations, we extended the neural network RandLA-Net. In addition to the evaluation of RandLA-Net in this context, we also provide a reference dataset consisting of semantically annotated hyperspectral 3D point clouds. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG. |
Beschreibung: | Cited by: 0; Conference name: 17th International Conference on Intelligent Autonomous Systems, IAS-17; Conference date: 13 June 2022 through 16 June 2022; Conference code: 290269 |
ISBN: | 9783031222153 | ISSN: | 2367-3370 | DOI: | 10.1007/978-3-031-22216-0_21 | Externe URL: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85148765465&doi=10.1007%2f978-3-031-22216-0_21&partnerID=40&md5=72dc23f5a70cf3a17b0817a8ce872ee1 |
Zur Langanzeige
Seitenaufrufe
17
Letzte Woche
0
0
Letzter Monat
4
4
geprüft am 07.06.2024