OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environments
IEEE Transactions on Image Processing · 2025 · 34 · 3096-3107
DOI: 10.1109/tip.2025.3567828 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Chubin Zhang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 25-mal zitiert (Crossref) und nennt 94 Literaturangaben. Gefördert von: National Natural Science Foundation of China, the Guangdong Natural Science Funds for Distinguished Young Scholar.
Abstract (Originalsprache)
Occupancy prediction reconstructs 3D structures of surrounding environments. It provides detailed information for autonomous driving planning and navigation. However, most existing methods heavily rely on the LiDAR point clouds to generate occupancy ground truth, which is not available in the vision-based system. In this paper, we propose an OccNeRF method for training occupancy networks without 3D ground truth. Different from previous works which consider a bounded scene, we parameterize the reconstructed occupancy fields and reorganize the sampling strategy to align with the cameras' infinite perceptive range. The neural rendering is adopted to convert occupancy fields to multi-camera depth maps, supervised by multi-frame photometric consistency. Moreover, for semantic occupancy prediction, we design several strategies to polish the prompts and filter the outputs of a pretrained open-vocabulary 2D segmentation model. Extensive experiments for both self-supervised depth estimation and 3D occupancy prediction tasks on nuScenes and SemanticKITTI datasets demonstrate the effectiveness of our method. The code is available at https://github.com/LinShan-Bin/OccNeRF.
Förderung
- National Natural Science Foundation of China
- the Guangdong Natural Science Funds for Distinguished Young Scholar
Zitieren
APA
Zhang, C., Yan, J., Wei, Y., Li, J., Liu, L., Tang, Y., Duan, Y., & Lu, J. (2025). OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environments. IEEE Transactions on Image Processing, 34, 3096-3107. https://doi.org/10.1109/tip.2025.3567828
Vancouver
Zhang C, Yan J, Wei Y, Li J, Liu L, Tang Y, et al. OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environments. IEEE Transactions on Image Processing. 2025;34:3096-3107. doi:10.1109/tip.2025.3567828
BibTeX
@article{Zhang2025,
author = {Zhang, Chubin and Yan, Juncheng and Wei, Yi and Li, Jiaxin and Liu, Li and Tang, Yansong and Duan, Yueqi and Lu, Jiwen},
title = {OccNeRF: Advancing 3D Occupancy Prediction in LiDAR-Free Environments},
journal = {IEEE Transactions on Image Processing},
year = {2025},
volume = {34},
pages = {3096--3107},
doi = {10.1109/tip.2025.3567828}
}
EUVIMED