EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LiGaussOcc: Fully Self-Supervised 3D Semantic Occupancy Prediction from LiDAR via Gaussian Splatting

Zhiqiang Wei, Tao Huang iD, Fengdeng Zhang

Sensors · 20.09.2025 · 25 (18) · 5889

DOI: 10.3390/s25185889 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 20.09.2025. Verfasst von Zhiqiang Wei und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate 3D semantic occupancy perception is critical for autonomous driving, enabling robust navigation in unstructured environments. While vision-based methods suffer from depth inaccuracies and lighting sensitivity, LiDAR-based approaches face challenges due to sparse data and dependence on expensive manual annotations. This work proposes LiGaussOcc, a novel self-supervised framework for dense LiDAR-based 3D semantic occupancy prediction. Our method first encodes LiDAR point clouds into voxel features and addresses sparsity via an Empty Voxel Inpainting (EVI) module, refined by an Adaptive Feature Fusion (AFF) module. During training, a Gaussian Primitive from Voxels (GPV) module generates parameters for 3D Gaussian Splatting, enabling efficient rendering of 2D depth and semantic maps. Supervision is achieved through photometric consistency across adjacent camera views and pseudo-labels from vision–language models, eliminating manual 3D annotations. Evaluated on the nuScenes-OpenOccupancy benchmark, LiGaussOcc achieved performance competitive with 30.4% Intersection over Union (IoU) and 14.1% mean Intersection over Union (mIoU). It reached 91.6% of the performance of the fully supervised LiDAR-based L-CONet, while completely eliminating the need for costly and labor-intensive manual 3D annotations. It excelled particularly in static environmental classes, such as drivable surfaces and man-made structures. This work presents a scalable, annotation-free solution for LiDAR-based 3D semantic occupancy perception.

Schlagwörter

Stichwörter der Autor:innen

3D semantic occupancyGaussian renderingLiDAR perceptionautonomous drivingpoint cloudself-supervisionvoxelization

Autor:innen und Institutionen

  • School of Optical-Electrical and Computer Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.
  • College of Science and Engineering, James Cook University, Cairns, QLD 4878, Australia.

ORCID-Profile: Tao Huang

Zitieren

APA

Wei, Z., Huang, T., & Zhang, F. (2025). LiGaussOcc: Fully Self-Supervised 3D Semantic Occupancy Prediction from LiDAR via Gaussian Splatting. Sensors, 25(18), 5889. https://doi.org/10.3390/s25185889

Vancouver

Wei Z, Huang T, Zhang F. LiGaussOcc: Fully Self-Supervised 3D Semantic Occupancy Prediction from LiDAR via Gaussian Splatting. Sensors. 2025;25(18):5889. doi:10.3390/s25185889

BibTeX
@article{Wei2025,
  author = {Wei, Zhiqiang and Huang, Tao and Zhang, Fengdeng},
  title = {LiGaussOcc: Fully Self-Supervised 3D Semantic Occupancy Prediction from LiDAR via Gaussian Splatting},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {18},
  pages = {5889},
  doi = {10.3390/s25185889}
}
Kostenlos starten →