Point Cloud Semantic Segmentation Based on Rotation Invari-ance and Feature Aggregation

Xiujuan Liang; Shiye Zhang

doi:10.62051/ijcsit.v8n1.06

Authors

Xiujuan Liang
Shiye Zhang

DOI:

https://doi.org/10.62051/ijcsit.v8n1.06

Keywords:

3D point cloud, Point cloud semantic segmentation, Attention mechanism, Global information, Feature learning

Abstract

As a key task in 3D scene understanding, point cloud semantic segmentation has broad application prospects in fields such as autonomous driving and robot navigation. Existing point cloud segmentation methods suffer from insufficient local feature extraction and a lack of effective integration of global contextual information, leading to inaccurate recogni-tion and incomplete segmentation of categories with similar surface textures and geometric structures. In view of this, this paper proposes an improved point cloud segmentation method for RandLA-Net : (1) Local polar coordinate posi-tion encoding module is introduced to eliminate the impact of Z-axis rotation on feature learning; (2) Global information acquisition module composed of attention mechanisms is constructed to enhance the network's contextual perception ability; (3) Hybrid pooling mechanism is integrated to improve the extraction of local features. The proposed method is evaluated on the self-built HPU dataset and public datasets S3DIS and Toronto-3D. The results show that the improved network achieves mean intersection over union (mIoU) values of 90.7%, 71.2%, and 76.4% respectively, demonstrating improvements compared to other algorithms. The model exhibits excellent generalization and segmentation perfor-mance in different types of point cloud scenes.

Downloads

Download data is not yet available.

References

[1] Ding, Z.; Sun, Y.; Xu, S.; Pan, Y.; Peng, Y.; Mao, Z. Recent Advances and Perspectives in Deep Learning Techniques for 3D Point Cloud Data Processing. Robotics 2023, 12, 100, doi:10.3390/robotics12040100.

[2] Yurtsever, E.; Lambert, J.; Carballo, A.; Takeda, K. A Survey of Autonomous Driving: Common Practices and Emerging Technologies. IEEE Access 2020, 8, 58443–58469, doi:10.1109/ACCESS.2020.2983149.

[3] Ning, D.; Huang, S. L-PCM: Localization and Point Cloud Registration-Based Method for Pose Calibration of Mo-bile Robots. Information 2024, 15, 269, doi:10.3390/info15050269.

[4] Lyu, B.; Wang, Y. Immersive Visualization of 3D Subsurface Ground Model Developed from Sparse Boreholes Us-ing Virtual Reality (VR). Underground Space 2024, 17, 188–206, doi:10.1016/j.undsp.2023.11.004.

[5] Khairnar, S.; Thepade, S.D.; Kolekar, S.; Gite, S.; Pradhan, B.; Alamri, A.; Patil, B.; Dahake, S.; Gaikwad, R.; Chaudhari, A. Enhancing Semantic Segmentation for Autonomous Vehicle Scene Understanding in Indian Con-text Using Modified CANet Model. Methodsx 2025, 14, 103131, doi:10.1016/j.mex.2024.103131.

[6] Wang, W.; Tan, X.; Li, L.; Liu, Y.; Chang, Q. 3D-NLM: Voxel-Based Non-Local Means for 3D Point Cloud Noise De-tection and Smoothing. Comput. Graphics 2025, 132, 104348, doi:10.1016/j.cag.2025.104348.

[7] Singh, D.P.; Yadav, M. Deep Learning-Based Semantic Segmentation of Three-Dimensional Point Cloud: A Com-prehensive Review. Int. J. Remote Sens. 2024, 45, 532–586, doi:10.1080/01431161.2023.2297177.

[8] Guo, Z.; Zhang, Y.; Zhu, L.; Wang, H.; Jiang, G. TSC-PCAC: Voxel Transformer and Sparse Convolution-Based Point Cloud Attribute Compression for 3D Broadcasting. IEEE Trans. Broadcast. 2025, 71, 154–166, doi:10.1109/TBC.2024.3464417.

[9] Li, Y.; Li, Q.; Gao, C.; Gao, S.; Wu, H.; Liu, R. PFENet: Towards Precise Feature Extraction from Sparse Point Cloud for 3D Object Detection. Neural Networks 2025, 185, 107144, doi:10.1016/j.neunet.2025.107144.

[10] Zeng, Z.; Xu, Y.; Xie, Z.; Tang, W.; Wan, J.; Wu, W. Large-Scale Point Cloud Semantic Segmentation via Local Per-ception and Global Descriptor Vector. Expert Syst. Appl. 2024, 246, 123269, doi:10.1016/j.eswa.2024.123269.

[11] Liu, Q.; Yuan, H.; Su, H.; Liu, H.; Wang, Y.; Yang, H.; Hou, J. PQA-Net: Deep No Reference Point Cloud Quality Assessment via Multi-View Projection. IEEE Trans. Circuits Syst. Video Technol. 2021, 31, 4645–4660, doi:10.1109/TCSVT.2021.3100282.

[12] Wang, Z.; Yin, M.; Dong, J.; Zheng, H.; Ou, D.; Xie, L.; Yin, G. Multi-View Point Clouds Registration Method Based on Overlap-Area Features and Local Distance Constraints for the Optical Measurement of Blade Profiles. IEEE/ASME Trans. Mechatronics 2022, 27, 2729–2739, doi:10.1109/TMECH.2021.3119435.

[13] Yu, H.-T.; Song, M. MM-Point: Multi-View Information-Enhanced Multi-Modal Self-Supervised 3D Point Cloud Understanding. Proc. AAAI Conf. Artif. Intell. 2024, 38, 6773–6781, doi:10.1609/aaai.v38i7.28501.

[14] Charles, R.Q.; Hao, S.; Mo, K.; Guibas, L.J. PointNet: Deep Learning on Point Sets for 3D Classification and Seg-mentation. In Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); IEEE: Honolulu, HI, July 2017; pp. 77–85.

[15] Qi, C.R.; Yi, L.; Su, H.; Guibas, L.J. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space 2017.

[16] Li, Y.; Bu, R.; Sun, M.; Wu, W.; Di, X.; Chen, B. PointCNN: Convolution on X-Transformed Points. In Proceedings of the Advances in Neural Information Processing Systems; Curran Associates, Inc., 2018; Vol. 31.

[17] Wu, W.; Qi, Z.; Fuxin, L. PointConv: Deep Convolutional Networks on 3D Point Clouds.; 2019; pp. 9621–9630.

[18] Thomas, H.; Qi, C.R.; Deschaud, J.-E.; Marcotegui, B.; Goulette, F.; Guibas, L.J. KPConv: Flexible and Deformable Convolution for Point Clouds 2019.

[19] Zhao, H.; Jiang, L.; Fu, C.-W.; Jia, J. PointWeb: Enhancing Local Neighborhood Features for Point Cloud Processing. In Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR); June 2019; pp. 5560–5568.

[20] Zhang, Z.; Hua, B.-S.; Yeung, S.-K. ShellNet: Efficient Point Cloud Convolutional Neural Networks Using Concen-tric Shells Statistics 2019.

[21] Hu, Q.; Yang, B.; Xie, L.; Rosa, S.; Guo, Y.; Wang, Z.; Trigoni, N.; Markham, A. RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds 2020.

[22] Ding, X.; Chen, H.; Zhang, X.; Han, J.; Ding, G. RepMLPNet: Hierarchical Vision MLP with Re-Parameterized Lo-cality 2022.

[23] Fan, S.; Dong, Q.; Zhu, F.; Lv, Y.; Ye, P.; Wang, F.-Y. SCF-Net: Learning Spatial Contextual Features for Large-Scale Point Cloud Segmentation.; 2021; pp. 14504–14513.

[24] Armeni, I.; Sener, O.; Zamir, A.R.; Jiang, H.; Brilakis, I.; Fischer, M.; Savarese, S. 3D Semantic Parsing of Large-Scale Indoor Spaces. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); June 2016; pp. 1534–1543.

[25] Tan, W.; Qin, N.; Ma, L.; Li, Y.; Du, J.; Cai, G.; Yang, K.; Li, J. Toronto-3D: A Large-Scale Mobile LiDAR Dataset for Semantic Segmentation of Urban Roadways.; 2020; pp. 202–203.

[26] Su, Y.; Liu, W.; Yuan, Z.; Cheng, M.; Zhang, Z.; Shen, X.; Wang, C. DLA-Net: Learning Dual Local Attention Fea-tures for Semantic Segmentation of Large-Scale Building Facade Point Clouds. Pattern Recognit. 2022, 123, 108372, doi:10.1016/j.patcog.2021.108372.

[27] Qiu, S.; Anwar, S.; Barnes, N. Semantic Segmentation for Real Point Cloud Scenes via Bilateral Augmentation and Adaptive Fusion 2021.

[28] Landrieu, L.; Simonovsky, M. Large-Scale Point Cloud Semantic Segmentation with Superpoint Graphs 2018.

[29] Wang, Y.; Sun, Y.; Liu, Z.; Sarma, S.E.; Bronstein, M.M.; Solomon, J.M. Dynamic Graph CNN for Learning on Point Clouds. Acm T. Graphic. 2019, 38, 1–12, doi:10.1145/3326362.

[30] Ma, L.; Li, Y.; Li, J.; Tan, W.; Yu, Y.; Chapman, M.A. Multi-Scale Point-Wise Convolutional Neural Networks for 3D Object Segmentation from LiDAR Point Clouds in Large-Scale Environments. IEEE Trans. Intell. Transp. Syst. 2021, 22, 821–836, doi:10.1109/TITS.2019.2961060.