{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T15:38:22Z","timestamp":1778081902485,"version":"3.51.4"},"reference-count":56,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"Horizon 2020 Innovation Project MediaVerse","award":["957252"],"award-info":[{"award-number":["957252"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2024]]},"DOI":"10.1109\/access.2024.3436585","type":"journal-article","created":{"date-parts":[[2024,8,1]],"date-time":"2024-08-01T18:19:51Z","timestamp":1722536391000},"page":"129705-129716","source":"Crossref","is-referenced-by-count":6,"title":["The Visual Saliency Transformer Goes Temporal: TempVST for Video Saliency Prediction"],"prefix":"10.1109","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-7348-9613","authenticated-orcid":false,"given":"Nikos","family":"Lazaridis","sequence":"first","affiliation":[{"name":"Centre for Research & Technology Hellas, Thessaloniki, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kostas","family":"Georgiadis","sequence":"additional","affiliation":[{"name":"Centre for Research & Technology Hellas, Thessaloniki, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fotis","family":"Kalaganis","sequence":"additional","affiliation":[{"name":"Centre for Research & Technology Hellas, Thessaloniki, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2297-4802","authenticated-orcid":false,"given":"Giorgos","family":"Kordopatis-Zilos","sequence":"additional","affiliation":[{"name":"VRG, FEE, Czech Technical University in Prague, Prague, Czech Republic"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5441-7341","authenticated-orcid":false,"given":"Symeon","family":"Papadopoulos","sequence":"additional","affiliation":[{"name":"Centre for Research & Technology Hellas, Thessaloniki, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1367-5133","authenticated-orcid":false,"given":"Spiros","family":"Nikolopoulos","sequence":"additional","affiliation":[{"name":"Centre for Research & Technology Hellas, Thessaloniki, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6447-9020","authenticated-orcid":false,"given":"Ioannis","family":"Kompatsiaris","sequence":"additional","affiliation":[{"name":"Centre for Research & Technology Hellas, Thessaloniki, Greece"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2013.2282897"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.23919\/ELMAR.2018.8534631"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/3177745"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2019.2928144"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/s11042-022-12442-w"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/s10489-022-03917-2"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-57687-9_5"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/s11042-019-08535-8"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1007\/11861898_32"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01264-9_37"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/34.730558"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2006.86"},{"key":"ref13","article-title":"Predicting human gaze using low-level saliency combined with face detection","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"20","author":"Cerf"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2006.10.001"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1167\/13.4.11"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/5.726791"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.358"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.38"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2017.2710620"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.71"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2017.2777665"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICIP.2016.7532629"},{"key":"ref24","article-title":"Recurrent mixture density network for spatiotemporal visual attention","volume-title":"Proc. 5th Int. Conf. Learn. Represent. (ICLR)","author":"Bazzani"},{"key":"ref25","article-title":"Predicting video saliency with object-to-motion CNN and two-layer convolutional LSTM","author":"Jiang","year":"2017","journal-title":"arXiv:1709.06316"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.510"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2022.04.080"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-81638-4_25"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3505244"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00468"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00060"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2019.2924417"},{"key":"ref34","article-title":"Simple vs complex temporal recurrences for video saliency prediction","volume-title":"Proc. Brit. Mach. Vis. Conf. (BMVC)","author":"Linardos"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2019.2936112"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.6927"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2020.107615"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01267-0_19"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00248"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-021-01519-y"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/IROS51168.2021.9635989"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ROBOT.2008.4543329"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/IROS.2011.6095124"},{"key":"ref44","article-title":"Temporal\u2013spatial feature pyramid for video saliency detection","author":"Chang","year":"2021","journal-title":"arXiv:2105.04213"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2021.3139743"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2018.2815601"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/s11042-024-19404-4"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2022.3172971"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref50","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"Proc. ICLR","author":"Dosovitskiy"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2102.05095"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58558-7_25"},{"key":"ref53","first-page":"1","article-title":"SalGAN: Visual saliency prediction with adversarial networks","volume-title":"Proc. CVPR Scene Understand. Workshop (SUNw)","author":"Pan"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2017.2787612"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33018521"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/7503.003.0073"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10380310\/10620181.pdf?arnumber=10620181","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,9,25]],"date-time":"2024-09-25T05:37:50Z","timestamp":1727242670000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10620181\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":56,"URL":"https:\/\/doi.org\/10.1109\/access.2024.3436585","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024]]}}}