Smart City Gnosys
Smart city article details
| Title | Psnet: Patch-Based Self-Attention Network For 3D Point Cloud Semantic Segmentation |
|---|---|
| ID_Doc | 43636 |
| Authors | Yi H.; Liu Y.; Wang M. |
| Year | 2025 |
| Published | Remote Sensing, 17, 12 |
| DOI | http://dx.doi.org/10.3390/rs17122012 |
| Abstract | LiDAR-captured 3D point clouds are widely used in self-driving cars and smart cities. Point-based semantic segmentation methods allow for more efficient use of the rich geometric information contained in 3D point clouds, so it has gradually replaced other methods as the mainstream deep learning method in 3D point cloud semantic segmentation. However, existing methods suffer from limited receptive fields and feature misalignment due to hierarchical downsampling. To address these challenges, we propose PSNet, a novel patch-based self-attention network that significantly expands the receptive field while ensuring feature alignment through a patch-aggregation paradigm. PSNet combines patch-based self-attention feature extraction with common point feature aggregation (CPFA) to implicitly model large-scale spatial relationships. The framework first divides the point cloud into overlapping patches to extract local features via multi-head self-attention, then aggregates features of common points across patches to capture long-range context. Extensive experiments on Toronto-3D and Complex Scene Point Cloud (CSPC) datasets validate PSNet’s state-of-the-art performance, achieving overall accuracies (OAs) of 98.4% and 97.2%, respectively, with significant improvements in challenging categories (e.g., +32.1% IoU for fences). Experimental results on the S3DIS dataset show that PSNet attains competitive mIoU accuracy (71.2%) while maintaining lower inference latency (7.03 s). The PSNet architecture achieves a larger receptive field coverage, which represents a significant advantage over existing methods. This work not only reveals the mechanism of patch-based self-attention for receptive field enhancement but also provides insights into attention-based 3D geometric learning and semantic segmentation architectures. Furthermore, it provides substantial references for applications in autonomous vehicle navigation and smart city infrastructure management. © 2025 by the authors. |
| Author Keywords | 3D point cloud; patch-based self-attention; receptive field; semantic segmentation |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
34830  | 0.901 | Li Y.; Ye Z.; Huang X.; HeLi Y.; Shuang F. | Lcl_Fda: Local Context Learning And Full-Level Decoder Aggregation Network For Large-Scale Point Cloud Semantic Segmentation | Neurocomputing, 621 (2025) |
| 48325 | 0.888 | Zhang L.; Zhang K. | Semi-Supervised Semantic Segmentation Network For Point Clouds Based On 3D Shape | Applied Sciences (Switzerland), 13, 6 (2023) |
| 17074 | 0.883 | Zaboli M.; Rastiveis H.; Hosseiny B.; Shokri D.; Sarasua W.A.; Homayouni S. | D-Net: A Density-Based Convolutional Neural Network For Mobile Lidar Point Clouds Classification In Urban Areas | Remote Sensing, 15, 9 (2023) |
| 18322 | 0.882 | Luo Z.; Zeng Z.; Tang W.; Wan J.; Xie Z.; Xu Y. | Dense Dual-Branch Cross Attention Network For Semantic Segmentation Of Large-Scale Point Clouds | IEEE Transactions on Geoscience and Remote Sensing, 62 (2024) |
| 18256 | 0.88 | Li Z.; Ning X.; Lv Z.; Shi Z.; Jin H.; Wang Y.; Zhou W. | Demf-Net: Dual-Branch Feature Enhancement And Multi-Scale Fusion For Semantic Segmentation Of Large-Scale Point Clouds; [Demf-Net：基于双分支增强和多尺度融合的大规模点云语义分割] | Journal of Graphics, 46, 2 (2025) |
| 13435 | 0.872 | Wu H.; Deng J.; Wen C.; Li X.; Wang C.; Li J. | Casa: A Cascade Attention Network For 3-D Object Detection From Lidar Point Clouds | IEEE Transactions on Geoscience and Remote Sensing, 60 (2022) |
| 16869 | 0.871 | Gao L.; Liu Y.; Chen X.; Liu Y.; Yan S.; Zhang M. | Cus3D: A New Comprehensive Urban-Scale Semantic-Segmentation 3D Benchmark Dataset | Remote Sensing, 16, 6 (2024) |
| 42261 | 0.871 | Zhang S.; Wang B.; Chen Y.; Zhang S.; Zhang W. | Point And Voxel Cross Perception With Lightweight Cosformer For Large-Scale Point Cloud Semantic Segmentation | International Journal of Applied Earth Observation and Geoinformation, 131 (2024) |
| 42264 | 0.867 | Deng C.; Peng Z.; Chen Z.; Chen R. | Point Cloud Deep Learning Network Based On Balanced Sampling And Hybrid Pooling | Sensors, 23, 2 (2023) |
| 61547 | 0.867 | Gu K.; Lv Z.; Zhu W.; Yi Y.; Jia F.; Ma J. | Weakly Supervised Point Cloud Semantic Segmentation Based On Multidimensional Feature Fusion And Feature Representation | 2023 10th International Forum on Electrical Engineering and Automation, IFEEA 2023 (2023) |