EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LSCA-RCNN: Large-Kernel Spatial Residual and Cascade Attention Network for Voxel-Based 3D Object Detection

Yuyang Liu, Zhanyuan Jiang iD, Min Mao, Kun Zhang, Yu Xu iD, Mingchen Zhu, Xianjun Wu

Sensors · 27.06.2026 · 26 (13) · 4089

DOI: 10.3390/s26134089 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 27.06.2026. Verfasst von Yuyang Liu und 6 weiteren Autor:innen. Beteiligte Einrichtung: College of Physics and Electronic Engineering, Xinyang Normal University, Xinyang 464000, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Xinyang Normal University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

LiDAR-based 3D object detection remains challenging due to sparse and irregular point cloud distributions, which degrade detection accuracy for small and occluded objects. In view of this, this paper proposes a novel two-stage voxel-based 3D detector, namely LSCA-RCNN, to address these issues. First, spatial residual blocks (SRBs) and large-kernel spatial-wise convolutions are integrated into the 3D backbone to suppress feature degradation and to expand the receptive fields for stable multi-scale feature learning. Second, a ConvNeXt-based 2D backbone with spatial attention is constructed to enhance discriminative feature representation of small objects. Third, a cascaded detection head embedded with fine-grained grouped convolutions and cross-stage cross-attention is designed to achieve progressive bounding box refinement and to improve localization precision. Extensive evaluations on the KITTI dataset with the R40 metric show that the proposed method achieves consistent performance improvements over the baseline. In the moderate setting, LSCA-RCNN increases the 3D AP by 2.12%, 7.66%, and 5.43% for cars, pedestrians, and cyclists, respectively, while achieving gains of 1.62%, 5.05%, and 7.05% under the hard setting. These results validate the effectiveness and robustness of the proposed LSCA-RCNN for complex and challenging autonomous driving detection tasks.

Schlagwörter

Stichwörter der Autor:innen

3D object detectionConvNeXt-based 2D backbonecascade detection headlidar point cloudspatial residual blocks (SRBs)spatial-wise convolutions

Autor:innen und Institutionen

  • College of Physics and Electronic Engineering, Xinyang Normal University, Xinyang 464000, China.
  • School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu 611731, China.
  • Xinyang Gumai Optronics Co., Ltd., Xinyang 464000, China.
  • School of Geomatics Science and Technology, Nanjing Tech University, Nanjing 210009, China.
  • Xinyang Central Semiconductor Technology Co., Ltd., Xinyang 464000, China.

ORCID-Profile: Zhanyuan Jiang, Yu Xu

Förderung

  • Xinyang Normal University

Zitieren

APA

Liu, Y., Jiang, Z., Mao, M., Zhang, K., Xu, Y., Zhu, M., & Wu, X. (2026). LSCA-RCNN: Large-Kernel Spatial Residual and Cascade Attention Network for Voxel-Based 3D Object Detection. Sensors, 26(13), 4089. https://doi.org/10.3390/s26134089

Vancouver

Liu Y, Jiang Z, Mao M, Zhang K, Xu Y, Zhu M, et al. LSCA-RCNN: Large-Kernel Spatial Residual and Cascade Attention Network for Voxel-Based 3D Object Detection. Sensors. 2026;26(13):4089. doi:10.3390/s26134089

BibTeX
@article{Liu2026,
  author = {Liu, Yuyang and Jiang, Zhanyuan and Mao, Min and Zhang, Kun and Xu, Yu and Zhu, Mingchen and Wu, Xianjun},
  title = {LSCA-RCNN: Large-Kernel Spatial Residual and Cascade Attention Network for Voxel-Based 3D Object Detection},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {13},
  pages = {4089},
  doi = {10.3390/s26134089}
}
Kostenlos starten →