Title
A comparative study of point clouds semantic segmentation using three different neural networks on the railway station dataset
Author
Lumban-Gaol, Y. A. (Geospatial Information Agency; Student TU Delft)
Chen, Z. (Student TU Delft)
Smit, M. (Student TU Delft)
Li, X. (Student TU Delft)
Erbaşu, M. A. (Student TU Delft)
Verbree, E. (TU Delft GIS Technologie; TU Delft Architecture and the Built Environment) 
Balado Frías, J. (TU Delft GIS Technologie)
Meijers, B.M. (TU Delft GIS Technologie)
van der Vaart, C.G. (ESRI)
Faculty
Architecture and the Built Environment
Date
2021
Abstract
Point cloud data have rich semantic representations and can benefit various applications towards a digital twin. However, they are unordered and anisotropically distributed, thus being unsuitable for a typical Convolutional Neural Networks (CNN) to handle. With the advance of deep learning, several neural networks claim to have solved the point cloud semantic segmentation problem. This paper evaluates three different neural networks for semantic segmentation of point clouds, namely PointNet++, PointCNN and DGCNN. A public indoor scene of the Amersfoort railway station is used as the study area. Unlike the typical indoor scenes and even more from the ubiquitous outdoor ones in currently available datasets, the station consists of objects such as the entrance gates, ticket machines, couches, and garbage cans. For the experiment, we use subsets from the data, remove the noise, evaluate the performance of the selected neural networks. The results indicate an overall accuracy of more than 90% for all the networks but vary in terms of mean class accuracy and mean Intersection over Union (IoU). The misclassification mainly occurs in the classes of couch and garbage can. Several factors that may contribute to the errors are analyzed, such as the quality of the data and the proportion of the number of points per class. The adaptability of the networks is also heavily dependent on the training location: the overall characteristics of the train station make a trained network for one location less suitable for another.
Subject
Deep learning
Indoor Scene
Point Clouds
Railway Station
Semantic Segmentation
To reference this document use:
http://resolver.tudelft.nl/uuid:4eadad2b-b7fc-45d9-aa7e-949d13ecb0a6
DOI
https://doi.org/10.5194/isprs-archives-XLIII-B3-2021-223-2021
ISSN
1682-1750
Source
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 43 (B3-2021), 223-228
Event
2021 24th ISPRS Congress Commission III: Imaging Today, Foreseeing Tomorrow, 2021-07-05 → 2021-07-09, Nice, France
Part of collection
Institutional Repository
Document type
journal article
Rights
© 2021 Y. A. Lumban-Gaol, Z. Chen, M. Smit, X. Li, M. A. Erbaşu, E. Verbree, J. Balado Frías, B.M. Meijers, C.G. van der Vaart