EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Local Contextual Attention for Enhancing Kernel Point Convolution in 3D Point Cloud Semantic Segmentation

Onur Can Bayrak, Melis Uzar

Applied Sciences · 29.08.2025 · 15 (17) · 9503

DOI: 10.3390/app15179503 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 29.08.2025. Verfasst von Onur Can Bayrak und 1 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 93 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Point cloud segmentation underpins various applications in geospatial analysis, such as autonomous navigation, urban planning, and management. Kernel Point Convolution (KPConv) has become a de facto standard for such tasks, yet its fixed geometric kernel limits the modeling of fine-grained contextual relationships—particularly in heterogeneous, real-world point cloud data. In this paper, we introduce the adaptation of a Local Contextual Attention (LCA) mechanism for the KPConv network, with reweighting kernel coefficients based on local feature similarity in the spatial proximity domain. Crucially, our lightweight design preserves KPConv’s distance-based weighting while embedding adaptive context aggregation, improving boundary delineation and small-object recognition without incurring significant computational or memory overhead. Our comprehensive experiments validate the efficacy of the proposed LCA block across multiple challenging benchmarks. Specifically, our method significantly improves segmentation performance by achieving a 20% increase in mean Intersection over Union (mIoU) on the STPLS3D dataset. Furthermore, we observe a 16% enhancement in mean F1 score (mF1) on the Hessigheim3D benchmark and a notable 15% improvement in mIoU on the Toronto3D dataset. These performance gains place LCA-KPConv among the top-performing methods reported in these benchmark evaluations. Trained models, prediction results, and the code of LCA are available in a GitHub opensource repository.

Zitieren

APA

Bayrak, O. C., & Uzar, M. (2025). Local Contextual Attention for Enhancing Kernel Point Convolution in 3D Point Cloud Semantic Segmentation. Applied Sciences, 15(17), 9503. https://doi.org/10.3390/app15179503

Vancouver

Bayrak OC, Uzar M. Local Contextual Attention for Enhancing Kernel Point Convolution in 3D Point Cloud Semantic Segmentation. Applied Sciences. 2025;15(17):9503. doi:10.3390/app15179503

BibTeX
@article{Bayrak2025,
  author = {Bayrak, Onur Can and Uzar, Melis},
  title = {Local Contextual Attention for Enhancing Kernel Point Convolution in 3D Point Cloud Semantic Segmentation},
  journal = {Applied Sciences},
  year = {2025},
  volume = {15},
  number = {17},
  pages = {9503},
  doi = {10.3390/app15179503}
}
Kostenlos starten →