基于对比学习和标签挖掘的点云分割算法  

Point Cloud Segmentation Algorithm Based on Contrastive Learning and Label Mining

作　　者：黄华[1,2,4] 卜一凡 许宏丽[1,3] 王晓荣 Huang Hua;Bu Yifan;Xu Hongli;Wang Xiaorong(School of Computer and Information Technology,Beijing Jiaotong University,Beijing 100044;Beijing Key Lab of Traffic Data Analysis and Mining(Beijing Jiaotong University),Beijing 100044;Key Laboratory of Beijing for Railway Engineering(Beijing Jiaotong University),Beijing 100044;Frontiers Science Center for Smart High-speed Railway System(Beijing Jiaotong University),Beijing 100044)

机构地区：[1]北京交通大学计算机与信息技术学院,北京100044 [2]交通数据分析与挖掘北京市重点实验室(北京交通大学),北京100044 [3]轨道工程北京市重点实验室(北京交通大学),北京100044 [4]智慧高铁系统前沿科学中心(北京交通大学),北京100044

出　　处：《计算机研究与发展》2025年第1期132-143,共12页Journal of Computer Research and Development

基　　金：国家自然科学基金项目(51827813);国家重点研发计划项目(2022YFB2603302);北京市教育委员会科技重大项目(KJZD20191000402);中央高校基本科研业务费专项资金(2022JBQY009)。

摘　　要：基于深度学习的点云分割算法通过设计复杂的特征提取模块,可以对高维空间点云进行有效的分割.但由于缺乏对边界点集的特征挖掘,使得其对边界分割的精度欠佳.已有将对比学习思想用于点云分割以解决边界区域分割性能不足问题的研究中,忽略了点云无序和稀疏特性,特征提取不够准确.对此,设计了基于对比学习和标签挖掘的点云分割模型CL2M(contrastive learning label mining),通过自注意力机制学习不同位置处点云更为精准的特征,并引入对比学习方法,提高了点云边界处的分割精度.在对比边界学习过程中通过深入挖掘语义空间中的标签并设计了基于标签分布的对比边界学习模块,使得高维空间点云标签分布包含更多的语义信息.CL2M充分利用标签的分布规律计算分布间的距离,可准确划分正负样本,减少了常规硬划分带来的累计错误.在2个公开数据集上进行的实验结果表明,CL2M在多个评价指标上优于既有的点云分割模型,验证了模型的有效性.Point cloud segmentation algorithm based on deep learning can effectively segment point clouds in high-dimensional space by designing complex feature extraction modules.However,the lack of feature mining for boundary point set results in suboptimal accuracy in boundary segmentation.Some studies have applied the idea of contrastive learning to point cloud segmentation to solve the problem of insufficient boundary region segmentation performance,but the disorder and sparse characteristics of point cloud have not been fully utilized,and the feature extraction is not accurate enough.To solve these problems,we propose CL2M to learn more accurate features of point clouds at different locations through the self-attention mechanism,and the contrastive learning method is introduced to improve the segmentation accuracy of point cloud boundaries.In the process of contrastive boundary learning,labels in semantic space are deeply mined and a contrastive boundary learning module based on label distribution is designed to make the label distribution of point cloud in high-dimensional space contain more semantic information.The model makes full use of the label distribution law to calculate the distance between distributions,and can accurately divide positive and negative samples,reducing the cumulative errors caused by conventional hard partition.The results on two public data sets show that CL2M is superior to the existing point cloud segmentation model on several evaluation indexes,which verifies the effectiveness of the model.

关 键 词：计算机视觉 点云分割 对比学习 自注意力机制 边界挖掘 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]