SIA OpenIR  > 机器人学研究室
Speeded up Low Rank Online Metric Learning for Object Tracking
Cong Y(丛杨); Fan BJ(范保杰); Liu J(刘霁); Luo JB(罗杰波); Yu HB(于海斌)
Department机器人学研究室
Source PublicationIEEE Transactions on Circuits and Systems for Video Technology
ISSN1051-8215
2015
Volume25Issue:6Pages:922-934
Indexed BySCI ; EI
EI Accession number20155201733939
WOS IDWOS:000357616000003
Contribution Rank1
Funding OrganizationNSFC (61105013,61375014,61203270); the foundation of Chinese Scholarship Council.
KeywordOnline Learning Metric Learning Semi-supervised Learning Low Rank Object Tracking
AbstractVisual object tracking can be considered as an online procedure to adaptively measure the foreground object similarity itself. However, many previous works usually adopt a fixed metric or offline metric learning to evaluate this dynamic process; even with some online metric learning trackers, their models often suffer from overfitting issues. To overcome these deficiencies, we propose a self-supervised tracking method that incorporates adaptive metric learning and semi-supervised learning into a unified framework. For similarity measurement, we design a new online metric learning model via low rank constraint to handle overfitting. Specially, we employ the max norm instead of the trace norm used in our previous work. This not only maintains the low rank property to overcome overfitting, but also reduces the computational complexity from O(n3) to O(n2), such that the new model is more suitable for object tracking. Moreover, by associating the information from stored training templates with unlabeled testing samples, a bi-linear graph is defined accordingly to propagate the label of each sample. High-confidence samples are then collected for self-training the model and updating the templates concurrently to handle large-scale. Experiments on various benchmark datasets and comparisons to several stateof- the-art methods demonstrate the effectiveness and efficiency of our algorithm. 
Language英语
WOS HeadingsScience & Technology ; Technology
WOS SubjectEngineering, Electrical & Electronic
WOS KeywordROBUST VISUAL TRACKING ; MODEL ; SIMILARITY
WOS Research AreaEngineering
Citation statistics
Cited Times:10[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/15472
Collection机器人学研究室
Corresponding AuthorCong Y(丛杨)
Affiliation1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
2.Department of Computer Science, University of Rochester, Rochester, NY, United States
3.College of Automation, Nanjing University of Posts and Telecommunications, Nanjing, China
4.Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
Recommended Citation
GB/T 7714
Cong Y,Fan BJ,Liu J,et al. Speeded up Low Rank Online Metric Learning for Object Tracking[J]. IEEE Transactions on Circuits and Systems for Video Technology,2015,25(6):922-934.
APA Cong Y,Fan BJ,Liu J,Luo JB,&Yu HB.(2015).Speeded up Low Rank Online Metric Learning for Object Tracking.IEEE Transactions on Circuits and Systems for Video Technology,25(6),922-934.
MLA Cong Y,et al."Speeded up Low Rank Online Metric Learning for Object Tracking".IEEE Transactions on Circuits and Systems for Video Technology 25.6(2015):922-934.
Files in This Item: Download All
File Name/Size DocType Version Access License
Speeded up Low Rank (6296KB)期刊论文出版稿开放获取ODC PDDLView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Cong Y(丛杨)]'s Articles
[Fan BJ(范保杰)]'s Articles
[Liu J(刘霁)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Cong Y(丛杨)]'s Articles
[Fan BJ(范保杰)]'s Articles
[Liu J(刘霁)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Cong Y(丛杨)]'s Articles
[Fan BJ(范保杰)]'s Articles
[Liu J(刘霁)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Speeded up Low Rank Online Metric Learning for Object Tracking.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.