《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (11): 3396-3402.DOI: 10.11772/j.issn.1001-9081.2022101552

• 人工智能 • 上一篇    

基于孪生自适应图卷积算法的点云分类与分割

李维刚1,2, 陈婷1(), 田志强1   

  1. 1.武汉科技大学 信息科学与工程学院,武汉 430081
    2.武汉科技大学 冶金自动化与检测技术教育部工程研究中心,武汉 430081
  • 收稿日期:2022-10-20 修回日期:2023-02-03 接受日期:2023-02-08 发布日期:2023-04-12 出版日期:2023-11-10
  • 通讯作者: 陈婷
  • 作者简介:李维刚(1977—),男,湖北咸宁人,教授,博士,主要研究方向:工业过程控制、人工智能、机器学习
    陈婷(1999—),女,湖北孝感人,硕士研究生,主要研究方向:深度学习、模式识别、点云数据处理 chenting_myself@163.com
    田志强(1996—),男,湖北武汉人,博士研究生,主要研究方向:计算机视觉。
  • 基金资助:
    湖北省重点研发计划项目(2020BAB098)

Point cloud classification and segmentation based on Siamese adaptive graph convolution algorithm

Weigang LI1,2, Ting CHEN1(), Zhiqiang TIAN1   

  1. 1.School of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan Hubei 430081,China
    2.Engineering Research Center for Metallurgical Automation and Measurement Technology of Ministry of Education;(Wuhan University of Science and Technology),Wuhan Hubei 430081,China
  • Received:2022-10-20 Revised:2023-02-03 Accepted:2023-02-08 Online:2023-04-12 Published:2023-11-10
  • Contact: Ting CHEN
  • About author:LI Weigang, born in 1977, Ph. D., professor. His research interests include industrial process control, artificial intelligence, machine learning.
    CHEN Ting, born in 1999, M. S. candidate. Her research interests include deep learning, pattern recognition, point cloud data processing.
    TIAN Zhiqiang, born in 1996, Ph. D. candidate. His research interests include computer vision.
  • Supported by:
    Key Research and Development Program of Hubei Province(2020BAB098)

摘要:

点云数据具有稀疏性、不规则性和置换不变性,缺乏拓扑信息,导致它的特征难以被提取,为此,提出一种孪生自适应图卷积算法(SAGCA)进行点云分类与分割。首先,构建特征关系图挖掘不规则、稀疏点云特征间的拓扑关系;其次,引入共享卷积学习权重的孪生构图思想,保证点云的置换不变性,使拓扑关系表达更准确;最后,采用整体、局部两种结合方式,将SAGCA与各种处理点云数据的深度学习网络相结合,增强网络的特征提取能力。分别在ScanObjectNN、ShapeNetPart和S3DIS数据集上进行分类、对象部件分割和场景语义分割实验的结果表明,相较于PointNet++基准网络,基于同样的数据集和评价标准,SAGCA分类实验的类别平均准确率(mAcc)提高了2.80个百分点,对象部件分割实验的总体类别平均交并比(IoU)提高了2.31个百分点,场景语义分割实验的类别平均交并比(mIoU)提高了2.40个百分点,说明SAGCA能有效增强网络的特征提取能力,适用于多种点云分类分割任务。

关键词: 点云数据, 拓扑关系, 孪生, 自适应图卷积, 分类, 分割

Abstract:

Point cloud data has sparsity, irregularity, and permutation invariance, and lacks topological information, which makes it difficult to extract features of point cloud. Therefore, a Siamese Adaptive Graph Convolution Algorithm (SAGCA) was proposed for point cloud classification and segmentation. Firstly, the topological relationships between irregular and sparse point cloud features were mined by constructing feature relationship graph. Then, the Siamese composition idea of sharing convolution learning weights was introduced to ensure the permutation invariance of point cloud data and make the topological relationship expression more accurate. Finally, SAGCA was combined with various deep learning networks for processing point cloud data by both global and local combination methods, thereby enhancing the feature extraction ability of the network. Comparison results with PointNet++ benchmark network of the classification, object part segmentation and scene semantic segmentation experiments on ScanObjectNN, ShapeNetPart and S3DIS datasets, respectively, show that, based on the same dataset and evaluation criteria, SAGCA has the class mean Accuracy (mAcc) of classification increased by 2.80 percentage points, the overall class average Intersection over Union (IoU) of part segmentation increased by 2.31 percentage points, and the class mean Intersection over Union (mIoU) of scene semantic segmentation increased by 2.40 percentage points, verifying that SAGCA can effectively enhance the feature extraction ability of the network and is suitable for multiple point cloud classification and segmentation tasks.

Key words: point cloud data, topological relationship, Siamese, adaptive graph convolution, classification, segmentation

中图分类号: