EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GeoSeqNet: A Geometry-Aware Sequential Network for Robust 3D Point Cloud Analysis

Dongzhen Liu iD, Yuzhong Deng iD, Haojie Wu, Jianxiao Zou, Shicai Fan

Sensors · 16.07.2026 · 26 (14) · 4511

DOI: 10.3390/s26144511 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 16.07.2026. Verfasst von Dongzhen Liu und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

3D point cloud understanding plays a vital role in remote sensing, robotic perception and intelligent scene analysis. However, real-world point cloud data are often affected by sensing noise, incomplete geometry, occlusion, and irregular sampling, posing significant challenges to reliable geometric representation learning and long-range contextual modeling. Existing methods typically rely on fixed neighborhood aggregation or computationally expensive global interaction mechanisms, leaving considerable room for improvement in terms of robustness and efficiency under complex sensing conditions. To address these challenges, we propose GeoSeqNet, a geometry-aware contextual learning framework for robust 3D point cloud analysis. Specifically, an Enhanced Local Operator (ELO) is introduced to strengthen local geometric representation, while a Geometric Encoding Module (GEM) is employed to preserve spatial geometric information during long-range feature interactions. In addition, an Adaptive Gate Fusion (AGF) module is designed to effectively integrate Gate-Scaled LSTM and GRU branches, enabling efficient long-range contextual modeling. By jointly exploiting local geometric cues and long-range contextual information, GeoSeqNet achieves robust feature learning with low computational overhead. Extensive experiments on ModelNet40, ScanObjectNN, and ShapeNetPart demonstrate the effectiveness of GeoSeqNet. The proposed method achieves competitive performance while maintaining a favorable efficiency–accuracy trade-off and exhibits strong robustness in complex real-world scenarios. These results indicate that GeoSeqNet provides an effective and reliable solution for point cloud understanding in challenging sensing environments.

Schlagwörter

Stichwörter der Autor:innen

3D point cloud analysisgeometry-aware learninglong-range contextual interactionrobust point cloud understandingsequential modeling

Autor:innen und Institutionen

  • School of Automation Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China.
  • Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China, Shenzhen 518110, China.

ORCID-Profile: Dongzhen Liu, Yuzhong Deng

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Liu, D., Deng, Y., Wu, H., Zou, J., & Fan, S. (2026). GeoSeqNet: A Geometry-Aware Sequential Network for Robust 3D Point Cloud Analysis. Sensors, 26(14), 4511. https://doi.org/10.3390/s26144511

Vancouver

Liu D, Deng Y, Wu H, Zou J, Fan S. GeoSeqNet: A Geometry-Aware Sequential Network for Robust 3D Point Cloud Analysis. Sensors. 2026;26(14):4511. doi:10.3390/s26144511

BibTeX
@article{Liu2026,
  author = {Liu, Dongzhen and Deng, Yuzhong and Wu, Haojie and Zou, Jianxiao and Fan, Shicai},
  title = {GeoSeqNet: A Geometry-Aware Sequential Network for Robust 3D Point Cloud Analysis},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {14},
  pages = {4511},
  doi = {10.3390/s26144511}
}
Kostenlos starten →