Online object tracking based on L1-loss SVMs with motion constraints

Tao Zhuo, Peng Zhang, Yanning Zhang, Wei Huang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Orange technologies focus on individual behavior analysis, and the core of which is object tracking, especially arbitrary object tracking. One of the popular solution for arbitrary object tracking is tracking by detection. These approaches regard the tracking problem as a detection task, and use the online learning methods to adapt the classifier to various object appearance changes. However, due to lack of prior knowledge and unpredictable appearance changes, it is always hard to get accurate target location during the whole tracking process. In this paper, we incorporate a motion model into the tracking by detection framework. Besides object prediction, the motion model also guides the model updating process to guarantee the performance of the classifier. Experimentally, we show that our algorithm is able to outperform state of art trackers on benchmark data sets.

Original languageEnglish
Title of host publicationIEEE International Conference on Orange Technologies, ICOT 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages61-64
Number of pages4
ISBN (Electronic)9781479962846
DOIs
StatePublished - 12 Nov 2014
Event2014 IEEE International Conference on Orange Technologies, ICOT 2014 - Xi'an, China
Duration: 20 Sep 201423 Sep 2014

Publication series

NameIEEE International Conference on Orange Technologies, ICOT 2014

Conference

Conference2014 IEEE International Conference on Orange Technologies, ICOT 2014
Country/TerritoryChina
CityXi'an
Period20/09/1423/09/14

Keywords

  • motion constraints
  • SVM
  • Tracking

Fingerprint

Dive into the research topics of 'Online object tracking based on L1-loss SVMs with motion constraints'. Together they form a unique fingerprint.

Cite this