Dual-Template Siamese Network with Attention Feature Fusion for Object Tracking

Simple item page
dc.contributor.authorLiu,M. H.
dc.contributor.authorShi,J. T.
dc.contributor.authorWang,Y.
dc.coverage.issue3cs
dc.coverage.volume32cs
dc.date.accessioned2023-10-11T08:00:48Z
dc.date.available2023-10-11T08:00:48Z
dc.date.issued2023-09cs
dc.description.abstractIn order to alleviate the adverse effects resulted from complex scenes for object tracking, such as fast movement, mottled background, interference of similar objects, and occlusion etc., an algorithm using dual-template Siamese network with attention feature fusion, named SiamDT, is proposed in this paper. The main idea include that the original ResNet-50 network is improved to extract deep semantic information and shallow spatial information, which are effectively fused using the attention mechanism to achieve accurate feature representation of objects. In addition, a template branch is added to the traditional Siamese network in which a dynamic template is generated together with the first frame image to solve the problems of template failure and model drift. Experimental results on OTB100 dataset and VOT2018 dataset show that the proposed approach obtains the excellent performance compared with the state-of-the-art tracking algorithms, which verifies the feasibility and effectiveness of the proposed approach.en
dc.formattextcs
dc.format.extent371-380cs
dc.format.mimetypeapplication/pdfen
dc.identifier.citationRadioengineering. 2023 vol. 32, č. 3, s. 371-380. ISSN 1210-2512cs
dc.identifier.doi10.13164/re.2023.0371en
dc.identifier.issn1210-2512
dc.identifier.urihttp://hdl.handle.net/11012/214345
dc.language.isoencs
dc.publisherSpolečnost pro radioelektronické inženýrstvícs
dc.relation.ispartofRadioengineeringcs
dc.relation.urihttps://www.radioeng.cz/fulltexts/2023/23_03_0371_0380.pdfcs
dc.rightsCreative Commons Attribution 4.0 International licenseen
dc.rights.accessopenAccessen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en
dc.subjectObject trackingen
dc.subjectSiamese networken
dc.subjectfeature extractionen
dc.subjectfeature fusionen
dc.subjectattention mechanismen
dc.titleDual-Template Siamese Network with Attention Feature Fusion for Object Trackingen
dc.type.driverarticleen
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
eprints.affiliatedInstitution.facultyFakulta eletrotechniky a komunikačních technologiícs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
23_03_0371_0380.pdf
Size:
1.12 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
2023/3