SIA OpenIR  > 机器人学研究室
Consistent multi-layer subtask tracker via hyper-graph regularization
Fan BJ(范保杰); Cong Y(丛杨)
Department机器人学研究室
Source PublicationPattern Recognition
ISSN0031-3203
2017
Volume67Pages:299-312
Indexed BySCI ; EI
EI Accession number20171303494215
WOS IDWOS:000399520700025
Contribution Rank2
Funding OrganizationChina Postdoctoral Science Foundation (No. 2015M571785, 2016T90484), NSFC (U1613214, 61673254, 61533015), Jiangsu Postdoctoral Science Foundation (No. 1402085C), State Key Laboratory of Robotics Open Project and the Foundation for the Talent of Nanjing University of Tele. and Com. (No. NY215148), project supported by the open fund of Key Laboratoryof Measurement and Control of Complex Systems of Engineering,Ministry of Education (No. MCCSE2015A05).
KeywordMulti-layer Subtask Learning Intrinsic Geometrical Structure Graph Regularization Normalized Collaborate Metric Object Tracking
AbstractMost multi-task learning based trackers adopt similar task definition by assuming that all tasks share a common feature set, which can't cover the real situation well. In this paper, we define the subtasks from the novel perspective, and develop a structured and consistent multi-layer multi-subtask tracker with graph regularization. The tracking task is completed by the collaboration of multi-layer subtasks. Different subtasks correspond to the tracking of different parts in the target area. The correspondences of the subtasks among the adjacent frames are consistent and smooth. The proposed model introduces hyper-graph regularizer to preserve the global and local intrinsic geometrical structures among and inside target candidates or trained samples, and decomposes the representative matrix of the subtasks into two components: low-rank property captures the subtask relationship, group-sparse property identifies the outlier subtasks. Moreover, a collaborate metric scheme is developed to find the best candidate, by concerning both discrimination reliability and representation accuracy. We show that the proposed multi-layer multi-subtask learning based tracker is a general model, which accommodates most existing multi-task trackers with the respective merits. Encouraging experimental results on a large set of public video sequences justify the effectiveness and robustness of the proposed tracker, and achieve comparable performance against many state-of-the-art methods.
Language英语
Citation statistics
Cited Times:1[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/21226
Collection机器人学研究室
Corresponding AuthorFan BJ(范保杰)
Affiliation1.Automation College, Nanjing University of Posts and Telecommunications, China
2.State Key Laboratory of Robotics, Chinese Academy of Sciences, China
Recommended Citation
GB/T 7714
Fan BJ,Cong Y. Consistent multi-layer subtask tracker via hyper-graph regularization[J]. Pattern Recognition,2017,67:299-312.
APA Fan BJ,&Cong Y.(2017).Consistent multi-layer subtask tracker via hyper-graph regularization.Pattern Recognition,67,299-312.
MLA Fan BJ,et al."Consistent multi-layer subtask tracker via hyper-graph regularization".Pattern Recognition 67(2017):299-312.
Files in This Item: Download All
File Name/Size DocType Version Access License
Consistent multi-lay(2648KB)期刊论文作者接受稿开放获取ODC PDDLView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Fan BJ(范保杰)]'s Articles
[Cong Y(丛杨)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Fan BJ(范保杰)]'s Articles
[Cong Y(丛杨)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Fan BJ(范保杰)]'s Articles
[Cong Y(丛杨)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Consistent multi-layer subtask tracker via hyper-graph regularization.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

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