Multi-template Scale-Adaptive Kernelized Correlation Filters

Handle URI:
http://hdl.handle.net/10754/605184
Title:
Multi-template Scale-Adaptive Kernelized Correlation Filters
Authors:
Bibi, Adel Aamer ( 0000-0002-6169-3918 ) ; Ghanem, Bernard ( 0000-0002-5534-587X )
Abstract:
This paper identifies the major drawbacks of a very computationally efficient and state-of-the-art-tracker known as the Kernelized Correlation Filter (KCF) tracker. These drawbacks include an assumed fixed scale of the target in every frame, as well as, a heuristic update strategy of the filter taps to incorporate historical tracking information (i.e. simple linear combination of taps from the previous frame). In our approach, we update the scale of the tracker by maximizing over the posterior distribution of a grid of scales. As for the filter update, we prove and show that it is possible to use all previous training examples to update the filter taps very efficiently using fixed-point optimization. We validate the efficacy of our approach on two tracking datasets, VOT2014 and VOT2015.
KAUST Department:
Electrical Engineering Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2015 IEEE International Conference on Computer Vision Workshop (ICCVW)
Conference/Event name:
2015 IEEE International Conference on Computer Vision Workshop (ICCVW)
Issue Date:
7-Dec-2015
DOI:
10.1109/ICCVW.2015.83
Type:
Conference Paper
Additional Links:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7406432
Appears in Collections:
Conference Papers; Electrical Engineering Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorBibi, Adel Aameren
dc.contributor.authorGhanem, Bernarden
dc.date.accessioned2016-04-13T13:31:04Zen
dc.date.available2016-04-13T13:31:04Zen
dc.date.issued2015-12-07en
dc.identifier.doi10.1109/ICCVW.2015.83en
dc.identifier.urihttp://hdl.handle.net/10754/605184en
dc.description.abstractThis paper identifies the major drawbacks of a very computationally efficient and state-of-the-art-tracker known as the Kernelized Correlation Filter (KCF) tracker. These drawbacks include an assumed fixed scale of the target in every frame, as well as, a heuristic update strategy of the filter taps to incorporate historical tracking information (i.e. simple linear combination of taps from the previous frame). In our approach, we update the scale of the tracker by maximizing over the posterior distribution of a grid of scales. As for the filter update, we prove and show that it is possible to use all previous training examples to update the filter taps very efficiently using fixed-point optimization. We validate the efficacy of our approach on two tracking datasets, VOT2014 and VOT2015.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.relation.urlhttp://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7406432en
dc.rights(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.en
dc.titleMulti-template Scale-Adaptive Kernelized Correlation Filtersen
dc.typeConference Paperen
dc.contributor.departmentElectrical Engineering Programen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journal2015 IEEE International Conference on Computer Vision Workshop (ICCVW)en
dc.conference.date7-13 Dec. 2015en
dc.conference.name2015 IEEE International Conference on Computer Vision Workshop (ICCVW)en
dc.conference.locationSantiagoen
dc.eprint.versionPost-printen
kaust.authorBibi, Adel Aameren
kaust.authorGhanem, Bernarden
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.