EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Point Cloud Classification and Segmentation Network Based on Adaptive Feature Extraction

Chengzhi Deng iD, Huaipei Wang, Zhaoming Wu, Xiaowei Sun, Shaoquan Zhang, Shengqian Wang

Sensors · 10.06.2026 · 26 (12) · 3689

DOI: 10.3390/s26123689 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 10.06.2026. Verfasst von Chengzhi Deng und 5 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Software, Zhaoqing University, Zhaoqing 526061, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, The Science and Technology Innovation Program of Jiangxi Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Point cloud classification and segmentation are key technologies for 3D perception and scene understanding, whose accuracy and efficiency directly affect the performance of high-level applications such as 3D modeling, object recognition, and intelligent interaction. Existing methods still exhibit obvious deficiencies in local feature representation, computational efficiency, and scene applicability. To address these issues, this paper proposes a lightweight point cloud classification and segmentation network based on adaptive feature extraction, referred to as AFE-PointNet. Firstly, an element-wise weighting set abstraction module based on the Hadamard product is designed. It leverages geometric topology learning to achieve adaptive feature enhancement, effectively improving the representation capability of local geometric structures. Meanwhile, a cascaded structure of feature aggregation and an inverted residual multi-layer perceptron (InvResMLP) is adopted for deep feature mining to achieve high-accuracy and high-efficiency point cloud classification and segmentation. Experimental results show that AFE-PointNet achieves an overall accuracy (OA) of 93.6% on the ModelNet40 dataset and 84.5% on the ScanObjectNN dataset, and attains a class mean intersection over union (Cls.mIoU) of 83.6% on the ShapeNetPart part segmentation dataset, yielding significant performance improvements over the PointNet++ model. The proposed adaptive feature enhancement and lightweight deep mining strategies effectively improve point cloud representation capability, providing a high-precision and efficient solution for 3D vision tasks.

Schlagwörter

Stichwörter der Autor:innen

Hadamard productadaptive feature extractioninverted residual MLPlightweight networkpoint cloud classificationpoint cloud segmentation

Autor:innen und Institutionen

  • School of Computer Science and Software, Zhaoqing University, Zhaoqing 526061, China.
  • Key Laboratory of Smart Water Conservancy in Jiangxi Province, Jiangxi University of Water Resources and Electric Power, Nanchang 330099, China.

ORCID-Profile: Chengzhi Deng

Förderung

  • National Natural Science Foundation of China
  • The Science and Technology Innovation Program of Jiangxi Province

Zitieren

APA

Deng, C., Wang, H., Wu, Z., Sun, X., Zhang, S., & Wang, S. (2026). Point Cloud Classification and Segmentation Network Based on Adaptive Feature Extraction. Sensors, 26(12), 3689. https://doi.org/10.3390/s26123689

Vancouver

Deng C, Wang H, Wu Z, Sun X, Zhang S, Wang S. Point Cloud Classification and Segmentation Network Based on Adaptive Feature Extraction. Sensors. 2026;26(12):3689. doi:10.3390/s26123689

BibTeX
@article{Deng2026,
  author = {Deng, Chengzhi and Wang, Huaipei and Wu, Zhaoming and Sun, Xiaowei and Zhang, Shaoquan and Wang, Shengqian},
  title = {Point Cloud Classification and Segmentation Network Based on Adaptive Feature Extraction},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {12},
  pages = {3689},
  doi = {10.3390/s26123689}
}
Kostenlos starten →