Towards Resource-Efficient Deep Learning for Train Scene Semantic Segmentation

Marie-Claire Iatrides, Marie-Claire Iatrides, Petra Gomez-Krämer, Olfa Ben Ahmed, Sylvain Marchand

2025

Abstract

In this paper, we present a promising application of scaling techniques for segmentation tasks in a railway environment context to highlight the advantages of task specific models tailored for on-board train use. Smaller convolutional neural networks (CNNs) do not focus on accuracy but resource efficiency. Our models are scaled using skip connections as well as quantization in order to form lightweight models trained specifically for our context. The proposed models have been evaluated both in terms of segmentation performance and efficiency on state of the art scene segmentation datasets namely RailSem19 and Cityscapes. We have obtained models with less than 3.5M parameters and a minimum of 78.4% of segmentation accuracy showing that lightweight models can effectively segment the railway surroundings.

Paper Citation

in Harvard Style

Iatrides M., Gomez-Krämer P., Ben Ahmed O. and Marchand S. (2025). Towards Resource-Efficient Deep Learning for Train Scene Semantic Segmentation. In Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP; ISBN 978-989-758-728-3, SciTePress, pages 347-354. DOI: 10.5220/0013139000003912

in Bibtex Style

@conference{visapp25,
author={Marie-Claire Iatrides and Petra Gomez-Krämer and Olfa Ben Ahmed and Sylvain Marchand},
title={Towards Resource-Efficient Deep Learning for Train Scene Semantic Segmentation},
booktitle={Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP},
year={2025},
pages={347-354},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013139000003912},
isbn={978-989-758-728-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 3: VISAPP
TI - Towards Resource-Efficient Deep Learning for Train Scene Semantic Segmentation
SN - 978-989-758-728-3
AU - Iatrides M.
AU - Gomez-Krämer P.
AU - Ben Ahmed O.
AU - Marchand S.
PY - 2025
SP - 347
EP - 354
DO - 10.5220/0013139000003912
PB - SciTePress