CASIA OpenIR  > 模式识别国家重点实验室  > 图像与视频分析
Part Context Learning for Visual Tracking
Zhu, Guibo; Wang, Jinqiao; Zhao, Chaoyang; Lu, Hanqing; Jinqiao Wang
Conference NameBritish Machine Computer Vision
Source PublicationIn proceedings of British Machine Computer Vision
Conference DateSeptember 1-5
Conference PlaceNottingham, UK
AbstractContext information is widely used in computer vision for tracking arbitrary objects. Most existing works focus on how to distinguish the tracked object from background or inter-frame object similarity information or key-points supporters as their auxiliary information to assist them in tracking. However, in most cases, how to discover and represent both the intrinsic property inside the object and surrounding information is still an open problem. In this paper, we propose a unified context learning framework that can capture stable structure relations of in-object parts, context parts and the object itself to enhance the tracker’s performance. The proposed Part Context Tracker (PCT) consists of an appearance model, an internal relation model and an context relation model. The appearance model represents the appearances of the object and parts. The internal relation model utilizes the parts inside the object to describe the spatio temporal structure property directly, while the context relation model takes advantage of the latent intersection between the object and background parts. Then the appearance model, internal relation model and context relation model are embedded in a max-margin structured learning framework. Furthermore, a simple robust update strategy using median filter is utilized, which can deal with appearance change effectively and alleviate the drift problem. Extensive experiments are conducted on various benchmark dataset, and the comparisons with state-of-the-arts demonstrate the effectiveness of our work.
KeywordVisual Tracking Part Context Learning
Indexed ByEI
Document Type会议论文
Corresponding AuthorJinqiao Wang
AffiliationChinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Zhu, Guibo,Wang, Jinqiao,Zhao, Chaoyang,et al. Part Context Learning for Visual Tracking[C],2014.
Files in This Item: Download All
File Name/Size DocType Version Access License
Part Context Learnin(1971KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhu, Guibo]'s Articles
[Wang, Jinqiao]'s Articles
[Zhao, Chaoyang]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhu, Guibo]'s Articles
[Wang, Jinqiao]'s Articles
[Zhao, Chaoyang]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhu, Guibo]'s Articles
[Wang, Jinqiao]'s Articles
[Zhao, Chaoyang]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Part Context Learning for Visual 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.