Visual Tracking Based on Convolutional Neural Network - 文章详情页

作者：

Li, Xiuzhi (Li, Xiuzhi.) | Jiang, Kai (Jiang, Kai.) | Jia, Songmin (Jia, Songmin.) (学者：贾松敏) | Zhang, Xiangyin (Zhang, Xiangyin.) | Sun, Yanjun (Sun, Yanjun.)

收录：

EI Scopus

摘要：

The　tracking　object　is　one　of　the　important　research　directions　in　the　field　of　computer　vision　and　plays　an　important　role　in　intelligent　video　monitoring.　In　this　paper,　a　visual　tracking　method　based　on　deep　learning　object　detection　is　proposed.　There　are　many　functions　required　in　human　object　tracking　tasks,　including　the　drive　of　hight-definition　panoramic　camera,　real-time　video　streaming　protocol　of　RTSP,　object　detection　based　on　deep　convolution　neural　network,　ROI　selection　of　interest　area,　dynamic　object　tracking　of　KF,　and　online　video　distribution　of　human　coordinates　through　SOCKET　communication.　©　2018　IEEE.

关键词：

Convolution Convolutional neural networks Deep learning Deep neural networks Neural networks Object detection Object recognition Object tracking

作者机构：

[ 1 ] [Li, Xiuzhi]Faculty of Information Technology, Beijing University of Technology, Beijing, China, China
[ 2 ] [Li, Xiuzhi]Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing, China
[ 3 ] [Jiang, Kai]Faculty of Information Technology, Beijing University of Technology, Beijing, China, China
[ 4 ] [Jiang, Kai]Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing, China
[ 5 ] [Jia, Songmin]Faculty of Information Technology, Beijing University of Technology, Beijing, China, China
[ 6 ] [Jia, Songmin]Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing, China
[ 7 ] [Zhang, Xiangyin]Faculty of Information Technology, Beijing University of Technology, Beijing, China, China
[ 8 ] [Zhang, Xiangyin]Beijing Key Laboratory of Computational Intelligence and Intelligent System, Beijing, China
[ 9 ] [Sun, Yanjun]Faculty of Information Technology, Beijing University of Technology, Beijing, China, China

通讯作者信息：

电子邮件地址：

查看成果更多字段

成果类型
所属机构

所有年份指定年份从至