Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-05 · 47 (5) · 3748-3765
DOI: 10.1109/tpami.2025.3535625 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025-05. Verfasst von Lingdong Kong und 7 weiteren Autor:innen. Die Arbeit wurde bisher 49-mal zitiert (Crossref) und nennt 111 Literaturangaben.
Abstract (Originalsprache)
Efficient data utilization is crucial for advancing 3D scene understanding in autonomous driving, where reliance on heavily human-annotated LiDAR point clouds challenges fully supervised methods. Addressing this, our study extends into semi-supervised learning for LiDAR semantic segmentation, leveraging the intrinsic spatial priors of driving scenes and multi-sensor complements to augment the efficacy of unlabeled datasets. We introduce LaserMix++, an evolved framework that integrates laser beam manipulations from disparate LiDAR scans and incorporates LiDAR-camera correspondences to further assist data-efficient learning. Our framework is tailored to enhance 3D scene consistency regularization by incorporating multi-modality, including 1) multi-modal LaserMix operation for fine-grained cross-sensor interactions; 2) camera-to-LiDAR feature distillation that enhances LiDAR feature learning; and 3) language-driven knowledge guidance generating auxiliary supervisions using open-vocabulary models. The versatility of LaserMix++ enables applications across LiDAR representations, establishing it as a universally applicable solution. Our framework is rigorously validated through theoretical analysis and extensive experiments on popular driving perception datasets. Results demonstrate that LaserMix++ markedly outperforms fully supervised alternatives, achieving comparable accuracy with five times fewer annotations and significantly improving the supervised-only baselines. This substantial advancement underscores the potential of semi-supervised approaches in reducing the reliance on extensive labeled data in LiDAR-based 3D scene understanding systems.
Zitieren
APA
Kong, L., Xu, X., Ren, J., Zhang, W., Pan, L., Chen, K., Ooi, W. T., & Liu, Z. (2025). Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving. IEEE Transactions on Pattern Analysis and Machine Intelligence, 47(5), 3748-3765. https://doi.org/10.1109/tpami.2025.3535625
Vancouver
Kong L, Xu X, Ren J, Zhang W, Pan L, Chen K, et al. Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2025;47(5):3748-3765. doi:10.1109/tpami.2025.3535625
BibTeX
@article{Kong2025,
author = {Kong, Lingdong and Xu, Xiang and Ren, Jiawei and Zhang, Wenwei and Pan, Liang and Chen, Kai and Ooi, Wei Tsang and Liu, Ziwei},
title = {Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2025},
volume = {47},
number = {5},
pages = {3748--3765},
doi = {10.1109/tpami.2025.3535625}
}
EUVIMED