Please use this identifier to cite or link to this item: https://open.uns.ac.rs/handle/123456789/6589
DC FieldValueLanguage
dc.contributor.authorRomani E.en
dc.contributor.authorda Silva W.en
dc.contributor.authorFonseca K.en
dc.contributor.authorĆulibrk, Dubravkoen
dc.contributor.authorPohl A.en
dc.date.accessioned2019-09-30T08:56:05Z-
dc.date.available2019-09-30T08:56:05Z-
dc.date.issued2015-01-01en
dc.identifier.isbn9783319232218en
dc.identifier.issn3029743en
dc.identifier.urihttps://open.uns.ac.rs/handle/123456789/6589-
dc.description.abstract© Springer International Publishing Switzerland 2015. This paper uses models of visual attention in order to estimate the human visual perception and thus improve metrics of Video Quality Assessment. This work reports on the use of the saliency based model in a fullreference structural similarity metric for creating new metrics that take into account regions that greatly attract the human attention. Correlation results with the differential mean opinion score values from the LIVE Video Quality Database are presented and discussed.en
dc.relation.ispartofLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)en
dc.titleFull-reference SSIM metric for video quality assessment with saliency-based featuresen
dc.typeConference Paperen
dc.identifier.doi10.1007/978-3-319-23222-5_66en
dc.identifier.scopus2-s2.0-84944680763en
dc.identifier.urlhttps://api.elsevier.com/content/abstract/scopus_id/84944680763en
dc.relation.lastpage554en
dc.relation.firstpage547en
dc.relation.volume9281en
item.grantfulltextnone-
item.fulltextNo Fulltext-
crisitem.author.deptFakultet tehničkih nauka, Departman za industrijsko inženjerstvo i menadžment-
crisitem.author.parentorgFakultet tehničkih nauka-
Appears in Collections:FTN Publikacije/Publications
Show simple item record

SCOPUSTM   
Citations

1
checked on May 10, 2024

Page view(s)

42
Last Week
20
Last month
0
checked on May 10, 2024

Google ScholarTM

Check

Altmetric


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