UPRIGHT ADJUSTMENT WITH GRAPH CONVOLUTIONAL NETWORKS

Cited 3 time in webofscience Cited 0 time in scopus
  • Hit : 172
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorJung, Raehyukko
dc.contributor.authorCho, Sungminko
dc.contributor.authorKwon, Junseokko
dc.date.accessioned2021-10-21T08:50:20Z-
dc.date.available2021-10-21T08:50:20Z-
dc.date.created2021-10-19-
dc.date.created2021-10-19-
dc.date.created2021-10-19-
dc.date.issued2020-09-
dc.identifier.citation2020 IEEE International Conference on Image Processing, ICIP 2020, pp.1058 - 1062-
dc.identifier.issn1522-4880-
dc.identifier.urihttp://hdl.handle.net/10203/288300-
dc.description.abstractWe present a novel method for the upright adjustment of 360 degrees. images. Our network consists of two modules, which are a convolutional neural network (CNN) and a graph convolutional network (GCN). The input 360 degrees. images is processed with the CNN for visual feature extraction, and the extracted feature map is converted into a graph that finds a spherical representation of the input. We also introduce a novel loss function to address the issue of discrete probability distributions defined on the surface of a sphere. Experimental results demonstrate that our method outperforms fully connected-based methods.-
dc.languageEnglish-
dc.publisherIEEE-
dc.titleUPRIGHT ADJUSTMENT WITH GRAPH CONVOLUTIONAL NETWORKS-
dc.typeConference-
dc.identifier.wosid000646178501032-
dc.identifier.scopusid2-s2.0-85098633453-
dc.type.rimsCONF-
dc.citation.beginningpage1058-
dc.citation.endingpage1062-
dc.citation.publicationname2020 IEEE International Conference on Image Processing, ICIP 2020-
dc.identifier.conferencecountryAR-
dc.identifier.conferencelocationVirtual-
dc.identifier.doi10.1109/ICIP40778.2020.9190715-
dc.contributor.localauthorJung, Raehyuk-
dc.contributor.nonIdAuthorCho, Sungmin-
dc.contributor.nonIdAuthorKwon, Junseok-
Appears in Collection
RIMS Conference Papers
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 3 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0