胡小鹏

个人信息Personal Information

教授

博士生导师

硕士生导师

性别:男

毕业院校:帝国理工学院

学位:博士

所在单位:计算机科学与技术学院

学科:计算机应用技术. 信号与信息处理

办公地点:创新园大厦-A0922

联系方式:18641135356

电子邮箱:xphu@dlut.edu.cn

扫描关注

论文成果

当前位置: 中文主页 >> 科学研究 >> 论文成果

Reliability verification-based convolutional neural networks for object tracking

点击次数:

论文类型:期刊论文

发表时间:2019-01-01

发表刊物:IET IMAGE PROCESSING

收录刊物:SCIE、Scopus

卷号:13

期号:1

页面范围:175-185

ISSN号:1751-9659

关键字:object tracking; learning (artificial intelligence); feedforward neural nets; reliability; feature selection; tracking network; reliability verification-based convolutional neural networks; object tracking; tracking algorithm; reliability analysis; deep network; single network; improperly labelled training samples; feature selection; label enhancement; convolutional layers

摘要:The authors propose a tracking algorithm based on the reliability analysis of the convolutional neural network to avoid drift. In general, most tracking algorithms implemented with the deep network consist of a single network; they obtain the tracking results according to the confidence and perform updates with the samples, which are collected based on the previous target state. However, this kind of algorithm relies heavily on the accuracy of tracking results, and slight deviations can lead to improperly labelled training samples and degrade the network. Therefore, they design a verification network to guarantee the reliability of the tracking network by correcting the results and it can be connected to a tracking network by sharing convolutional layers. The reliability verification network estimates the accuracy of the results of the tracking network and discards ambiguous results to avoid accumulating errors. Specifically, the verification network can distinguish the target from the confused candidates more precisely because of the optimised training data. The training samples of the verification network consist of characteristics and labels, and they are optimised by feature selection and label enhancement, respectively. The experimental results illustrate the outstanding performance compared with several state-of-the-art methods on the challenging video sequences.