EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Sparse-to-Dense Point Cloud Registration Based on Rotation-Invariant Features

Tianjiao Ma, Guangliang Han, Yongzhi Chu, Hong Ren

Remote Sensing · 06.07.2024 · 16 (13) · 2485

DOI: 10.3390/rs16132485 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 06.07.2024. Verfasst von Tianjiao Ma und 3 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 58 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Department of Science and Technology of Jilin Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Point cloud registration is a critical problem because it is the basis of many 3D vision tasks. With the popularity of deep learning, many scholars have focused on leveraging deep neural networks to address the point cloud registration problem. However, many of these methods are still sensitive to partial overlap and differences in density distribution. For this reason, herein, we propose a method based on rotation-invariant features and using a sparse-to-dense matching strategy for robust point cloud registration. Firstly, we encode raw points as superpoints with a network combining KPConv and FPN, and their associated features are extracted. Then point pair features of these superpoints are computed and embedded into the transformer to learn the hybrid features, which makes the approach invariant to rigid transformation. Subsequently, a sparse-to-dense matching strategy is designed to address the registration problem. The correspondences of superpoints are obtained via sparse matching and then propagated to local dense points and, further, to global dense points, the byproduct of which is a series of transformation parameters. Finally, the enhanced features based on spatial consistency are repeatedly fed into the sparse-to-dense matching module to rebuild reliable correspondence, and the optimal transformation parameter is re-estimated for final alignment. Our experiments show that, with the proposed method, the inlier ratio and registration recall are effectively improved, and the performance is better than that of other point cloud registration methods on 3DMatch and ModelNet40.

Förderung

  • Department of Science and Technology of Jilin Province

Zitieren

APA

Ma, T., Han, G., Chu, Y., & Ren, H. (2024). Sparse-to-Dense Point Cloud Registration Based on Rotation-Invariant Features. Remote Sensing, 16(13), 2485. https://doi.org/10.3390/rs16132485

Vancouver

Ma T, Han G, Chu Y, Ren H. Sparse-to-Dense Point Cloud Registration Based on Rotation-Invariant Features. Remote Sensing. 2024;16(13):2485. doi:10.3390/rs16132485

BibTeX
@article{Ma2024,
  author = {Ma, Tianjiao and Han, Guangliang and Chu, Yongzhi and Ren, Hong},
  title = {Sparse-to-Dense Point Cloud Registration Based on Rotation-Invariant Features},
  journal = {Remote Sensing},
  year = {2024},
  volume = {16},
  number = {13},
  pages = {2485},
  doi = {10.3390/rs16132485}
}
Kostenlos starten →