Please use this identifier to cite or link to this item: https://repositorio.fei.edu.br/handle/FEI/1010
Full metadata record
DC FieldValueLanguage
dc.contributor.authorLOPES, Guilherme-
dc.contributor.authorHORVATH, M.-
dc.contributor.authorGIRALDI, G. A.-
dc.contributor.authorLOPES, Guilherme-
dc.date.accessioned2019-08-17T20:00:30Z-
dc.date.available2019-08-17T20:00:30Z-
dc.date.issued2019-
dc.identifier.citationLOPES, Guilherme; HORVATH, M.; GIRALDI, G. A.; LOPES, Guilherme. A strategy based on non-extensive statistics to improve frame-matching algorithms under large viewpoint changes. Signal Processing: Image Communication, v. 75, p. 44-54, 2019.-
dc.identifier.issn0923-5965-
dc.identifier.urihttps://repositorio.fei.edu.br/handle/FEI/1010-
dc.relation.ispartofSignal Processing: Image Communication-
dc.rightsAcesso Aberto-
dc.titleA strategy based on non-extensive statistics to improve frame-matching algorithms under large viewpoint changespt_BR
dc.typeArtigopt_BR
dc.identifier.doi10.1016/j.image.2019.03.007-
dc.description.volume75-
dc.description.firstpage44-
dc.description.lastpage54-
dc.identifier.urlhttps://doi.org/10.1016/j.image.2019.03.007-
dc.description.abstractalternativeIn recent decades, methods to find invariant points in digital images, called fiducial points, have gained greatattention, mainly due to the demands of several applications in computer vision and image processing, suchas the geometric matching of global structures, objects or specific regions. Among the most well knownapproaches are algorithms like SIFT, HOG, SURF, and their variations as A-SIFT, PCA-SIFT, surrounded bymany others. Though the number of researches demonstrating the efficiency of such methods is undoubtedlyenormous, the vast majority compares their performances only on pairs of images with little changes in viewperspectives, objects or specific regions of the scenes. Thereby, the study of this type of technique under largeviewpoint changes, called here LVC, has received little attention from the researchers. On the other hand, withthe aim of filtering points of interest, most techniques have used the traditional extensive statistics. However,methods for image processing based on a new type of statistics, called non-extensive statistics, have shownto be efficient in several applications. In this paper, we present a new method, called𝑞-SIFT, based on thenon-extensive Tsallis statistics, to find fiducial points in a sequence of frames of videos under large viewpointschanges. We experimentally show the efficiency of the proposed method in video databases and propose newmeasurement metrics for this type of algorithm.en
dc.subject.otherlanguageSIFTen
dc.subject.otherlanguageTsallis statisticsen
dc.subject.otherlanguage𝑞-Gaussian functionen
dc.subject.otherlanguageImage matchingen
dc.subject.otherlanguageImage registrationen
Appears in Collections:Artigos

Files in This Item:
There are no files associated with this item.


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