dc.contributor.author | Kasneci, Enkelejda | |
dc.contributor.author | Fuhl, Wolfgang | |
dc.contributor.author | Schneider, Johannes | |
dc.date.accessioned | 2022-11-09T13:01:36Z | |
dc.date.available | 2022-11-09T13:01:36Z | |
dc.date.issued | 2021-10-11 | |
dc.identifier.isbn | 978-1-6654-0191-3 | |
dc.identifier.uri | http://hdl.handle.net/10900/132936 | |
dc.language.iso | en | de_DE |
dc.publisher | IEEE | de_DE |
dc.relation.uri | https://doi.org/10.1109/ICCVW54120.2021.00386 | de_DE |
dc.subject.ddc | 004 | de_DE |
dc.title | 1000 Pupil Segmentations in a Second using Haar Like Features and Statistical Learning | de_DE |
dc.type | Article | de_DE |
dc.type | ConferenceObject | de_DE |
utue.personen.roh | Kasneci, Enkelejda | |
utue.personen.roh | Fuhl, Wolfgang | |
utue.personen.roh | Schneider, Johannes | |
dcterms.isPartOf.ZSTitelID | 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW) | de_DE |
Dateien | Größe | Format | Anzeige |
---|---|---|---|
Zu diesem Dokument gibt es keine Dateien. |