LiDAR as a Geometric Prior: Enhancing Camera Pose Tracking Through High-Fidelity View Synthesis
Applied Sciences · 07.08.2025 · 15 (15) · 8743
DOI: 10.3390/app15158743 ↗Kurz gefasst
Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 07.08.2025. Verfasst von Rafael Muñoz-Salinas und 4 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Science, Innovation and Universities of Spain, Hong Kong Research Grants Council (UGC) General Research Fund (GRF).
Abstract (Originalsprache)
This paper presents a robust framework for monocular camera pose estimation by leveraging high-fidelity, pre-built 3D LiDAR maps. The core of our approach is a render-and-match pipeline that synthesizes photorealistic views from a dense LiDAR point cloud. By detecting and matching keypoints between these synthetic images and the live camera feed, we establish reliable 3D–2D correspondences for accurate pose estimation. We evaluate two distinct strategies: an Online Rendering and Tracking method that renders views on the fly, and an Offline Keypoint-Map Tracking method that precomputes a keypoint map for known trajectories, optimizing for computational efficiency. Comprehensive experiments demonstrate that our framework significantly outperforms several state-of-the-art visual SLAM systems in both accuracy and tracking consistency. By anchoring localization to the stable geometric information from the LiDAR map, our method overcomes the reliance on photometric consistency that often causes failures in purely image-based systems, proving particularly effective in challenging real-world environments.
Förderung
- Ministry of Science, Innovation and Universities of Spain
- Hong Kong Research Grants Council (UGC) General Research Fund (GRF)
Zitieren
APA
Muñoz-Salinas, R., Liu, J., Romero-Ramirez, F. J., Marín-Jiménez, M. J., & Zhang, F. (2025). LiDAR as a Geometric Prior: Enhancing Camera Pose Tracking Through High-Fidelity View Synthesis. Applied Sciences, 15(15), 8743. https://doi.org/10.3390/app15158743
Vancouver
Muñoz-Salinas R, Liu J, Romero-Ramirez FJ, Marín-Jiménez MJ, Zhang F. LiDAR as a Geometric Prior: Enhancing Camera Pose Tracking Through High-Fidelity View Synthesis. Applied Sciences. 2025;15(15):8743. doi:10.3390/app15158743
BibTeX
@article{MuozSalinas2025,
author = {Muñoz-Salinas, Rafael and Liu, Jianheng and Romero-Ramirez, Francisco J. and Marín-Jiménez, Manuel J. and Zhang, Fu},
title = {LiDAR as a Geometric Prior: Enhancing Camera Pose Tracking Through High-Fidelity View Synthesis},
journal = {Applied Sciences},
year = {2025},
volume = {15},
number = {15},
pages = {8743},
doi = {10.3390/app15158743}
}
EUVIMED