Chen Siyuan, Fang Zhiwei, Wan Siyao, Zhou Ting, Chen Chunlin, Wang Meng, Li Qianming
School of Information Science and Engineering, Hunan Institute of Science and Technology, Yueyang, 414006, China.
Hunan Institute of Science and Technology, College of Mechanical Engineering, Yueyang, 414006, China.
Sci Rep. 2025 May 13;15(1):16545. doi: 10.1038/s41598-025-00789-7.
With the increasing use of 3D point cloud data in autonomous driving, robotic perception, and remote sensing, efficient and accurate point cloud analysis remains a critical challenge. This study presents PointGA, a lightweight Transformer-based model that enhances geometric perception for improved feature extraction and representation. First, PointGA expands the original 3D coordinates into various geometric information, introducing more prior knowledge into the network. Second, a trigonometric position encoding suitable for point clouds is designed, which effectively enhances the expressive capability of positional information and performs preliminary feature extraction through pooling layers, significantly improving the model's robustness across various tasks. Finally, a positional differential self-attention (PDA) mechanism with linear complexity is developed to optimize feature representation and achieve efficient computation. Experimental results demonstrate that PointGA achieves 87.6% overall accuracy on the ScanObjectNN dataset for classification and 66.2% mean intersection over union(mIoU) on the S3DIS Area 5 dataset for segmentation, outperforming existing methods. These findings highlight the model's capability to balance efficiency and accuracy, offering a promising solution for point cloud analysis tasks.
随着三维点云数据在自动驾驶、机器人感知和遥感领域的应用日益广泛，高效准确的点云分析仍然是一项严峻挑战。本研究提出了PointGA，这是一种基于Transformer的轻量级模型，可增强几何感知能力，以改进特征提取和表示。首先，PointGA将原始三维坐标扩展为各种几何信息，将更多先验知识引入网络。其次，设计了一种适用于点云的三角位置编码，有效增强了位置信息的表达能力，并通过池化层进行初步特征提取，显著提高了模型在各种任务中的鲁棒性。最后，开发了一种具有线性复杂度的位置差分自注意力（PDA）机制，以优化特征表示并实现高效计算。实验结果表明，PointGA在ScanObjectNN数据集上的分类总体准确率达到87.6%，在S3DIS区域5数据集上的分割平均交并比（mIoU）达到66.2%，优于现有方法。这些发现突出了该模型在平衡效率和准确性方面的能力，为点云分析任务提供了一个很有前景的解决方案。
