MetaSSC: Enhancing 3D semantic scene completion for autonomous driving through meta-learning and long-sequence modeling
Communications in Transportation Research · 2025-12 · 5 · 100184
DOI: 10.1016/j.commtr.2025.100184 ↗Kurz gefasst
Fachartikel in der Zeitschrift Communications in Transportation Research, 2025-12. Verfasst von Yansong Qu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Research Foundation of Korea.
Abstract (Originalsprache)
Semantic scene completion (SSC) is essential for achieving comprehensive perception in autonomous driving systems. However, existing SSC methods often overlook the high deployment costs in real-world applications. Traditional architectures, such as 3D Convolutional Neural Networks (3D CNNs) and self-attention mechanisms, face challenges in efficiently capturing long-range dependencies within 3D voxel grids, limiting their effectiveness. To address these issues, we introduce MetaSSC, a novel meta-learning-based framework for SSC that leverages deformable convolution, large-kernel attention, and the Mamba (D-LKA-M) model. Our approach begins with a voxel-based semantic segmentation (SS) pretraining task, aimed at exploring the semantics and geometry of incomplete regions while acquiring transferable meta-knowledge. Using simulated cooperative perception datasets, we supervise the perception training of a single vehicle using aggregated sensor data from multiple nearby connected autonomous vehicles (CAVs), generating richer and more comprehensive labels. This meta-knowledge is then adapted to the target domain through a dual-phase training strategy that does not add extra model parameters, enabling efficient deployment. To further enhance the model's capability in capturing long-sequence relationships within 3D voxel grids, we integrate Mamba blocks with deformable convolution and large-kernel attention into the backbone network. Extensive experiments demonstrate that MetaSSC achieves state-of-the-art performance, significantly outperforming competing models while also reducing deployment costs.
Förderung
- National Research Foundation of Korea
Zitieren
APA
Qu, Y., Xu, Z., Huang, Z., Sheng, Z., Chen, S., & Chen, T. (2025). MetaSSC: Enhancing 3D semantic scene completion for autonomous driving through meta-learning and long-sequence modeling. Communications in Transportation Research, 5, 100184. https://doi.org/10.1016/j.commtr.2025.100184
Vancouver
Qu Y, Xu Z, Huang Z, Sheng Z, Chen S, Chen T. MetaSSC: Enhancing 3D semantic scene completion for autonomous driving through meta-learning and long-sequence modeling. Communications in Transportation Research. 2025;5:100184. doi:10.1016/j.commtr.2025.100184
BibTeX
@article{Qu2025,
author = {Qu, Yansong and Xu, Zixuan and Huang, Zilin and Sheng, Zihao and Chen, Sikai and Chen, Tiantian},
title = {MetaSSC: Enhancing 3D semantic scene completion for autonomous driving through meta-learning and long-sequence modeling},
journal = {Communications in Transportation Research},
year = {2025},
volume = {5},
pages = {100184},
doi = {10.1016/j.commtr.2025.100184}
}
EUVIMED