A Novel Vision Transformer for Camera-LiDAR Fusion Based Traffic Object Segmentation

Toomas Tahves, Junyi Gu, Junyi Gu, Mauro Bellone, Raivo Sell

2025

Abstract

This paper presents Camera-LiDAR Fusion Transformer (CLFT) models for traffic object segmentation, which leverage the fusion of camera and LiDAR data using vision transformers. Building on the methodology of visual transformers that exploit the self-attention mechanism, we extend segmentation capabilities with additional classification options to a diverse class of objects including cyclists, traffic signs, and pedestrians across diverse weather conditions. Despite good performance, the models face challenges under adverse conditions which underscores the need for further optimization to enhance performance in darkness and rain. In summary, the CLFT models offer a compelling solution for autonomous driving perception, advancing the state-of-the-art in multimodal fusion and object segmentation, with ongoing efforts required to address existing limitations and fully harness their potential in practical deployments.

Paper Citation

in Harvard Style

Tahves T., Gu J., Bellone M. and Sell R. (2025). A Novel Vision Transformer for Camera-LiDAR Fusion Based Traffic Object Segmentation. In Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART; ISBN 978-989-758-737-5, SciTePress, pages 566-573. DOI: 10.5220/0013239000003890

in Bibtex Style

@conference{icaart25,
author={Toomas Tahves and Junyi Gu and Mauro Bellone and Raivo Sell},
title={A Novel Vision Transformer for Camera-LiDAR Fusion Based Traffic Object Segmentation},
booktitle={Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART},
year={2025},
pages={566-573},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013239000003890},
isbn={978-989-758-737-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART
TI - A Novel Vision Transformer for Camera-LiDAR Fusion Based Traffic Object Segmentation
SN - 978-989-758-737-5
AU - Tahves T.
AU - Gu J.
AU - Bellone M.
AU - Sell R.
PY - 2025
SP - 566
EP - 573
DO - 10.5220/0013239000003890
PB - SciTePress