IPFA-Net: Important Points Feature Aggregating Net for Point Cloud Classification and Segmentation

作     者：Jingya WANG Yu ZHANG Bin ZHANG Jinxiang XIA Weidong WANG 

作者机构：School of Information and Software Engineering University of Electronic Science and Technology of China National Key Laboratory of Automotive Chassis Integration and Bionics Jilin University 

出 版 物：《Chinese Journal of Electronics》 (电子学报(英文))

年 卷 期：2024年

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the Sichuan Province Science and Technology Support Program (Grant No.2021YFQ0054) the open project fund of Intelligent Terminal Key Laboratory of Sichuan Province (2020-2021) (Grant No. SCITLAB-0012) 

摘      要：This paper focuses on the problems of point cloud deep neural networks in classification and segmentation tasks, including losing important information during down-sampling, ignoring relationships among points when extracting features, and network performance being susceptible to the sparsity of point cloud. To begin with, this paper proposes a farthest point sampling(FPS)-important points sampling(F-IPS) method for down-sampling, which can preserve important information of point clouds and maintain the geometry of input data. Then, the local feature relation aggregating(LFRA) method is proposed for feature extraction, improving the network s ability to learn contextual information and extract rich local region features. Based on these methods, the important points feature aggregating net(IPFA-Net) is designed for point cloud classification and segmentation tasks. Furthermore, this paper proposes the multi-scale multi-density feature connecting(MMFC) method to reduce the negative impact of point cloud data sparsity on network performance. Finally, the effectiveness of IPFA-Net is demonstrated through experiments on ModelNet40, ShapeNet part, and ScanNet v2 datasets. IPFA-Net is robust to reducing the number of point clouds, with only a 3.3% decrease in accuracy under a 16-fold reduction of point number. In the part segmentation experiments, our method achieves the best segmentation performance for five objects.