Semantic segmentation of imbalanced 3D point clouds in railway environments : comparative analysis of algorithms and training pipelines for semantic segmentation

Tools

Ghasemlou, Arshia and Soilán, Mario and Riveiro, Belén; Moreno-Rangel, Alejandro and Kumar, Bimal, eds. (2025) Semantic segmentation of imbalanced 3D point clouds in railway environments : comparative analysis of algorithms and training pipelines for semantic segmentation. In: EG-ICE 2025. University of Strathclyde Publishing, GBR, pp. 547-557. ISBN 9781914241826 (https://doi.org/10.17868/strath.00093262)

[thumbnail of Ghasemlou-etal-EG-ICE-2025-Semantic-segmentation-of-imbalanced-3D-point-clouds]
Preview
 Text. Filename: Ghasemlou-etal-EG-ICE-2025-Semantic-segmentation-of-imbalanced-3D-point-clouds.pdf
Final Published Version
License: Creative Commons Attribution 4.0 logo
Download (1MB)| Preview

Abstract

Railway infrastructure is vital for modern transportation, and effective maintenance is crucial for ensuring safety and efficiency. Semantic segmentation of 3D point clouds can automate monitoring and support digital twins, yet it demands large amounts of training data. Existing large-scale datasets face challenges such as severe class imbalance, numerous unclassified points, and noise from distant objects, hindering model performance. In this paper, we address these issues using the SemanticRail3D dataset, which comprises 438 point clouds spanning 90 km of railway tracks with approximately 2.8 billion labeled points across 11 classes. We propose a novel preprocessing and training pipeline that employs PCA-based alignment, segmentation, a slight data augmentation, and tailored training configurations. Two state-of-the-art models, Swin3D and Point Transformer V3, are trained, achieving mean IoU scores of 82.47% and 82.95% on held-out test data, respectively. Our results demonstrate strong generalization and lay a solid foundation for automated railway infrastructure monitoring.

  • Item type:Book Section
    ID code:93262
    Dates:
    Date
    Event
    1 July 2025
    Published
    9 June 2025
    Accepted
    Subjects:Fine Arts > Architecture
    Department:Faculty of Engineering > Architecture
    Depositing user: Pure Administrator
    Date deposited:27 Jun 2025 10:48
    Last modified:01 Aug 2026 00:16
    URI:https://strathprints.strath.ac.uk/id/eprint/93262
CORE (COnnecting REpositories)