Adaptive Online Learning Based Robust Visual Tracking
Accurate location estimation of a target is a classical and very popular problem in visual object tracking, for which correlation filters have been proven highly effective in real-time scenarios. However, the great variation of the target's appearance and the surrounding background throughout a...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2018-01-01
|
Series: | IEEE Access |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/8308727/ |
id |
doaj-24bb8525b58c4dc7920ac584e8058a09 |
---|---|
record_format |
Article |
spelling |
doaj-24bb8525b58c4dc7920ac584e8058a092021-03-29T20:39:01ZengIEEEIEEE Access2169-35362018-01-016147901479810.1109/ACCESS.2018.28133748308727Adaptive Online Learning Based Robust Visual TrackingWeiming Yang0https://orcid.org/0000-0002-7101-670XMeirong Zhao1Yinguo Huang2Yelong Zheng3State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin, ChinaState Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin, ChinaState Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin, ChinaState Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, Tianjin, ChinaAccurate location estimation of a target is a classical and very popular problem in visual object tracking, for which correlation filters have been proven highly effective in real-time scenarios. However, the great variation of the target's appearance and the surrounding background throughout a video sequence would lead to failure tracking for the sake of the model drift, using trackers based on correlation filters. In our approach, we present a simple and fast method to improve the robustness of the model based on sum of template and pixel-wise learners (Staple). On the one hand, a confidence regression model is established to adjust adaptively the model online learning rate to alleviate the model drift. On the other hand, instead of likelihood, the scale with maximal posterior probability is selected as the target scale to obtain the more accurate estimation. Extensive experimental results demonstrate that the proposed approach performs favorably against several state-of-the-art algorithms on large-scale challenging benchmark data sets at speed in excess of 42 frames/s.https://ieeexplore.ieee.org/document/8308727/Adaptive online learningcorrelation filtersdiscriminative classifiervisual tracking |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Weiming Yang Meirong Zhao Yinguo Huang Yelong Zheng |
spellingShingle |
Weiming Yang Meirong Zhao Yinguo Huang Yelong Zheng Adaptive Online Learning Based Robust Visual Tracking IEEE Access Adaptive online learning correlation filters discriminative classifier visual tracking |
author_facet |
Weiming Yang Meirong Zhao Yinguo Huang Yelong Zheng |
author_sort |
Weiming Yang |
title |
Adaptive Online Learning Based Robust Visual Tracking |
title_short |
Adaptive Online Learning Based Robust Visual Tracking |
title_full |
Adaptive Online Learning Based Robust Visual Tracking |
title_fullStr |
Adaptive Online Learning Based Robust Visual Tracking |
title_full_unstemmed |
Adaptive Online Learning Based Robust Visual Tracking |
title_sort |
adaptive online learning based robust visual tracking |
publisher |
IEEE |
series |
IEEE Access |
issn |
2169-3536 |
publishDate |
2018-01-01 |
description |
Accurate location estimation of a target is a classical and very popular problem in visual object tracking, for which correlation filters have been proven highly effective in real-time scenarios. However, the great variation of the target's appearance and the surrounding background throughout a video sequence would lead to failure tracking for the sake of the model drift, using trackers based on correlation filters. In our approach, we present a simple and fast method to improve the robustness of the model based on sum of template and pixel-wise learners (Staple). On the one hand, a confidence regression model is established to adjust adaptively the model online learning rate to alleviate the model drift. On the other hand, instead of likelihood, the scale with maximal posterior probability is selected as the target scale to obtain the more accurate estimation. Extensive experimental results demonstrate that the proposed approach performs favorably against several state-of-the-art algorithms on large-scale challenging benchmark data sets at speed in excess of 42 frames/s. |
topic |
Adaptive online learning correlation filters discriminative classifier visual tracking |
url |
https://ieeexplore.ieee.org/document/8308727/ |
work_keys_str_mv |
AT weimingyang adaptiveonlinelearningbasedrobustvisualtracking AT meirongzhao adaptiveonlinelearningbasedrobustvisualtracking AT yinguohuang adaptiveonlinelearningbasedrobustvisualtracking AT yelongzheng adaptiveonlinelearningbasedrobustvisualtracking |
_version_ |
1724194335756386304 |