EUVIMEDEUVIMED entdecken
Crossref

Accurate and robust registration of low overlapping point clouds

Jieyin Yang, Mingyang Zhao, Yingrui Wu, Xiaohong Jia

Computers & Graphics · 2024-02 · 118 · 146-160

DOI: 10.1016/j.cag.2023.12.003 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers & Graphics, 2024-02. Verfasst von Jieyin Yang und 3 weiteren Autor:innen. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 68 Literaturangaben. Gefördert von: National Key Research and Development Program of China, Chinese Academy of Sciences, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Point cloud registration is fundamental to many computer graphics and visual tasks. Previous pairwise approaches mainly deal with the registration with a high overlapping hypothesis, while few existing methods explore the registration between low overlapping point clouds. However, the latter registration task is challenging and essential, since the weak correspondence in point clouds usually makes the algorithm tend to get stuck in a local minimum. In this paper, we present an accurate algorithm for the robust registration of low overlapping point clouds using optimal transformation. The core of our method is the effective integration of geometric features with the probabilistic model hidden Markov random field . First, we determine and remove the outliers of the point clouds by modeling a hidden Markov random field based on a high dimensional feature distribution. Then, we derive a necessary and sufficient condition when the symmetric function is minimized and present a new curvature-aware symmetric function to make the point correspondence more discriminative. Finally, we integrate our curvature-aware symmetric function into a geometrically stable sampling framework, which effectively constrains unstable transformations. We verify the accuracy and robustness of our method on a wide variety of datasets, particularly on low overlapping range scanned point clouds. Results demonstrate that our proposed method attains better performance with higher accuracy and robustness compared to representative state-of-the-art approaches.

Förderung

  • National Key Research and Development Program of China
  • Chinese Academy of Sciences
  • National Natural Science Foundation of China

Zitieren

APA

Yang, J., Zhao, M., Wu, Y., & Jia, X. (2024). Accurate and robust registration of low overlapping point clouds. Computers & Graphics, 118, 146-160. https://doi.org/10.1016/j.cag.2023.12.003

Vancouver

Yang J, Zhao M, Wu Y, Jia X. Accurate and robust registration of low overlapping point clouds. Computers & Graphics. 2024;118:146-160. doi:10.1016/j.cag.2023.12.003

BibTeX
@article{Yang2024,
  author = {Yang, Jieyin and Zhao, Mingyang and Wu, Yingrui and Jia, Xiaohong},
  title = {Accurate and robust registration of low overlapping point clouds},
  journal = {Computers & Graphics},
  year = {2024},
  volume = {118},
  pages = {146--160},
  doi = {10.1016/j.cag.2023.12.003}
}
Kostenlos starten →