Pillar-Based 3D Object Detection from Point Cloud with Multiattention Mechanism

Author:

Li Xin1ORCID,Liang Bifa1ORCID,Huang Jinhao1ORCID,Peng Yuyang2ORCID,Yan Yier1ORCID,Li Jun1ORCID,Shang Wenli1ORCID,Wei Wei1ORCID

Affiliation:

1. Research Center of Intelligent Communication Engineering, School of Electronics and Communication Engineering, Guangzhou University, Guangzhou 510006, China

2. School of Computer Science and Engineering, Macau University of Science and Technology, Taipa, 999078 Macau SAR, China

Abstract

Object detection in point clouds is a critical component in most autonomous driving systems. In this paper, in order to improve the effectiveness of image feature extraction and the accuracy of detection of point clouds, a pillar-based 3D point cloud object detection algorithm with multiattention mechanism is proposed, which includes three attention mechanisms SOCA, SOPA, and SAPI. The results show that the recognition accuracy of the optimized algorithm for cars, pedestrians, and cyclists on KITTI dataset is significantly improved on the detection benchmarks of BEV and 3D. Despite using only LiDAR, our algorithm outperforms PointPillars, which is one of the state-of-the-art algorithms for 3D object detection, with respect to both 3D and BEV view KITTI benchmarks while maintaining a relatively competitive speed.

Funder

International Collaborative Research Program of Guangdong Science and Technology Department

Publisher

Hindawi Limited

Subject

Electrical and Electronic Engineering,Computer Networks and Communications,Information Systems

Reference46 articles.

1. Machine Learning in Transportation

2. 5G NB-IoT: Efficient Network Traffic Filtering for Multitenant IoT Cellular Networks

3. Are Self-Driving Vehicles Ready to Launch? An Insight into Steering Control in Autonomous Self-Driving Vehicles

4. Voxelnet: end-to-end learning for point cloud based 3D object detection;Y. Zhou

5. Pixor: real-time 3D object detection from point clouds;B. Yang

Cited by 4 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3