EUVIMEDEUVIMED entdecken
Crossref

Online Targetless Camera–Lidar Extrinsic Calibration Using Object-Oriented Semantic Feature

Xinchun Ji, Zheng He, Shuhang Gu, Dongyan Wei, Deyun Zhou

IEEE Sensors Journal · 01.04.2026 · 26 (7) · 10526-10540

DOI: 10.1109/jsen.2026.3663294 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 01.04.2026. Verfasst von Xinchun Ji und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 30 Literaturangaben. Gefördert von: Open Fund of the Intelligent Control Laboratory, National Natural Science Foundation of China, Natural Science Foundation of Sichuan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate extrinsic parameters are critical for fusing heterogeneous data from camera and light detection and ranging (Lidar). To solve the problems of incomplete feature extraction, poor scene adaptability, and unstable parameter convergence in conventional automatic calibration methods, this article proposed an online extrinsic calibration method using object-oriented semantic features, which consisted of a coarse-calibration based on trajectory-consistency and a fine-calibration based on feature-consistency. With the initial parameters from the coarse-calibration, the fine-calibration pictorialized a sparse point cloud to obtain its depth, intensity, and class attributes simultaneously. Then, the segment anything model (SAM) network was used to process both the visual image and the point cloud pseudoimage to extract consistent object-oriented features. To enhance the stability and adaptability of extrinsic calibration to various scenes, an observability-based intersection-over-union (IoU) was proposed to perform the camera–Lidar feature matching. In addition, we designed an adaptability function to evaluate the quality of scene features, thereby ensuring the effectiveness of extrinsic calibration. Results of testing on the KITTI dataset and real-world dataset demonstrate that the proposed method can effectively obtain accurate extrinsic parameters in multiple scenes. By comparing with the ground truth, errors of the rotation and translation are less than (0.15°, 0.05 m), and the variances are less than (2.0 × 10-3 deg2, 3.0 × 10-4 m2), which indicates good potential for engineering applications.

Förderung

  • Open Fund of the Intelligent Control Laboratory
  • National Natural Science Foundation of China
  • Natural Science Foundation of Sichuan Province

Zitieren

APA

Ji, X., He, Z., Gu, S., Wei, D., & Zhou, D. (2026). Online Targetless Camera–Lidar Extrinsic Calibration Using Object-Oriented Semantic Feature. IEEE Sensors Journal, 26(7), 10526-10540. https://doi.org/10.1109/jsen.2026.3663294

Vancouver

Ji X, He Z, Gu S, Wei D, Zhou D. Online Targetless Camera–Lidar Extrinsic Calibration Using Object-Oriented Semantic Feature. IEEE Sensors Journal. 2026;26(7):10526-10540. doi:10.1109/jsen.2026.3663294

BibTeX
@article{Ji2026,
  author = {Ji, Xinchun and He, Zheng and Gu, Shuhang and Wei, Dongyan and Zhou, Deyun},
  title = {Online Targetless Camera–Lidar Extrinsic Calibration Using Object-Oriented Semantic Feature},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {7},
  pages = {10526--10540},
  doi = {10.1109/jsen.2026.3663294}
}
Kostenlos starten →