TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction
Semantic segmentation of point clouds plays a crucial role in computer vision, with diverse applications in urban modelling, autonomous driving, and virtual reality. Despite its significance, many existing methods face challenges when dealing with large-scale datasets, such as (1) unclear or incompl...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Geocarto International |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/10106049.2025.2489520 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850179495159398400 |
|---|---|
| author | Yong Li Falin Chen Qi Lin Zhen Li Dongxu Gao Jingchao Yang |
| author_facet | Yong Li Falin Chen Qi Lin Zhen Li Dongxu Gao Jingchao Yang |
| author_sort | Yong Li |
| collection | DOAJ |
| description | Semantic segmentation of point clouds plays a crucial role in computer vision, with diverse applications in urban modelling, autonomous driving, and virtual reality. Despite its significance, many existing methods face challenges when dealing with large-scale datasets, such as (1) unclear or incomplete boundary segmentation and (2) poor performance on sparse objects. These limitations stem from inadequate local context extraction and insufficient handling of density variations, which hinder the accuracy and robustness of segmentation. To address these challenges, we propose TFNet, an end-to-end deep neural network specifically designed to enhance local geometric feature extraction and improve performance on density variations. TFNet introduces three key components: (1) Rotation-Invariant and Geometric Feature Extractor (RIGFE), which independently captures rotation-invariant and geometric features; (2) Annularly Convolutional Attention Pooling (ACAP), which leverages annular convolution for effective relational feature extraction in both feature and geometric spaces; and (3) Subgraph Vector of Locally Aggregated Descriptors (SGVLAD), which learns position- and scale-invariant point set features. Experimental evaluations on benchmark datasets, including S3DIS, Toronto-3D, and Nanning Power Grid, demonstrate that TFNet outperforms existing methods by effectively addressing these challenges. The results highlight its ability to deliver superior segmentation accuracy and robustness in diverse scenarios. |
| format | Article |
| id | doaj-art-c2fa01301b244eacb815be3525966e1d |
| institution | OA Journals |
| issn | 1010-6049 1752-0762 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | Geocarto International |
| spelling | doaj-art-c2fa01301b244eacb815be3525966e1d2025-08-20T02:18:28ZengTaylor & Francis GroupGeocarto International1010-60491752-07622025-12-0140110.1080/10106049.2025.2489520TFNet: point cloud Semantic Segmentation Network based on Triple feature extractionYong Li0Falin Chen1Qi Lin2Zhen Li3Dongxu Gao4Jingchao Yang5Guangxi Key Laboratory of Power System Optimization and Energy Technology, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaGuangxi Key Laboratory of Power System Optimization and Energy Technology, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaGuangxi Key Laboratory of Power System Optimization and Energy Technology, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaGuangxi Key Laboratory of Power System Optimization and Energy Technology, School of Electrical Engineering, Guangxi University, Nanning 530004, ChinaSchool of Computing, University of Portsmouth, Portsmouth PO1 3HE, UKDepartment of Electrical and Information Engineering, Hebei Jiaotong Vocational and Technical College, Shijiazhuang, ChinaSemantic segmentation of point clouds plays a crucial role in computer vision, with diverse applications in urban modelling, autonomous driving, and virtual reality. Despite its significance, many existing methods face challenges when dealing with large-scale datasets, such as (1) unclear or incomplete boundary segmentation and (2) poor performance on sparse objects. These limitations stem from inadequate local context extraction and insufficient handling of density variations, which hinder the accuracy and robustness of segmentation. To address these challenges, we propose TFNet, an end-to-end deep neural network specifically designed to enhance local geometric feature extraction and improve performance on density variations. TFNet introduces three key components: (1) Rotation-Invariant and Geometric Feature Extractor (RIGFE), which independently captures rotation-invariant and geometric features; (2) Annularly Convolutional Attention Pooling (ACAP), which leverages annular convolution for effective relational feature extraction in both feature and geometric spaces; and (3) Subgraph Vector of Locally Aggregated Descriptors (SGVLAD), which learns position- and scale-invariant point set features. Experimental evaluations on benchmark datasets, including S3DIS, Toronto-3D, and Nanning Power Grid, demonstrate that TFNet outperforms existing methods by effectively addressing these challenges. The results highlight its ability to deliver superior segmentation accuracy and robustness in diverse scenarios.https://www.tandfonline.com/doi/10.1080/10106049.2025.2489520Large-scale point clouddeep learningpoint cloudsemantic segmentationlocal feature |
| spellingShingle | Yong Li Falin Chen Qi Lin Zhen Li Dongxu Gao Jingchao Yang TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction Geocarto International Large-scale point cloud deep learning point cloud semantic segmentation local feature |
| title | TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction |
| title_full | TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction |
| title_fullStr | TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction |
| title_full_unstemmed | TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction |
| title_short | TFNet: point cloud Semantic Segmentation Network based on Triple feature extraction |
| title_sort | tfnet point cloud semantic segmentation network based on triple feature extraction |
| topic | Large-scale point cloud deep learning point cloud semantic segmentation local feature |
| url | https://www.tandfonline.com/doi/10.1080/10106049.2025.2489520 |
| work_keys_str_mv | AT yongli tfnetpointcloudsemanticsegmentationnetworkbasedontriplefeatureextraction AT falinchen tfnetpointcloudsemanticsegmentationnetworkbasedontriplefeatureextraction AT qilin tfnetpointcloudsemanticsegmentationnetworkbasedontriplefeatureextraction AT zhenli tfnetpointcloudsemanticsegmentationnetworkbasedontriplefeatureextraction AT dongxugao tfnetpointcloudsemanticsegmentationnetworkbasedontriplefeatureextraction AT jingchaoyang tfnetpointcloudsemanticsegmentationnetworkbasedontriplefeatureextraction |