Semantic Encoding Supervises LiDAR Semantic Segmentation
IEEE Transactions on Intelligent Transportation Systems · 2025-10 · 26 (10) · 16973-16985
DOI: 10.1109/tits.2025.3569337 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Intelligent Transportation Systems, 2025-10. Verfasst von Chuanyu Luo und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 60 Literaturangaben. Gefördert von: LiangDao GmbH.
Abstract (Originalsprache)
Semantic segmentation using LiDAR is a fundamental aspect in perception for autonomous driving. Conventional training methodologies commonly employ one-hot encoding for labels, followed by a softmax activation and a cross-entropy loss function to facilitate model training. Nevertheless, this approach exhibits several limitations: 1) information loss due to the use of one-hot encoding, 2) overconfidence issues induced by the softmax function, and 3) inability to accurately predict unknown or unlabeled classes. In this paper, we propose to use a pre-trained semantic sentence encoding model for label encoding to address these shortages. We theoretically demonstrate that the proposed method maximizes the accuracy, mitigates the overconfidence, and enhances the mean intersection over union (mIoU) metric. Empirical experiments further validate our approach, showing that it not only improves the perception performance but also enables an effective zero-shot learning and a cross-dataset transfer with minimal trade-offs.
Förderung
- LiangDao GmbH
Zitieren
APA
Luo, C., Cheng, N., Chen, W., & Li, P. (2025). Semantic Encoding Supervises LiDAR Semantic Segmentation. IEEE Transactions on Intelligent Transportation Systems, 26(10), 16973-16985. https://doi.org/10.1109/tits.2025.3569337
Vancouver
Luo C, Cheng N, Chen W, Li P. Semantic Encoding Supervises LiDAR Semantic Segmentation. IEEE Transactions on Intelligent Transportation Systems. 2025;26(10):16973-16985. doi:10.1109/tits.2025.3569337
BibTeX
@article{Luo2025,
author = {Luo, Chuanyu and Cheng, Nuo and Chen, Wenyu and Li, Pu},
title = {Semantic Encoding Supervises LiDAR Semantic Segmentation},
journal = {IEEE Transactions on Intelligent Transportation Systems},
year = {2025},
volume = {26},
number = {10},
pages = {16973--16985},
doi = {10.1109/tits.2025.3569337}
}
EUVIMED