Show simple item record

dc.contributor.authorMokoena, Nthabiseng
dc.contributor.authorHelberg, Albert
dc.contributor.authorTsague, Hippolyte Djonon
dc.date.accessioned2017-05-09T13:00:20Z
dc.date.available2017-05-09T13:00:20Z
dc.date.issued2017
dc.identifier.citationMokoena, N. et al. 2017. 2D methods for pose invariant face recognition. 2016 International Conference on Computational Science and Computational Intelligence (CSCI), Las Vegas, NV, USA, 15-17 Dec. Article no 7881456:841-846. [http://ieeexplore.ieee.org/document/7881456/]en_US
dc.identifier.isbn978-1-5090-5510-4 (Online)
dc.identifier.urihttp://hdl.handle.net/10394/21811
dc.identifier.urihttp://dx.doi.org/10.1109/CSCI.2016.0163
dc.identifier.urihttp://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7881456
dc.description.abstractThe ability to recognise face images under random pose is a task that is done effortlessly by human beings. However, for a computer system, recognising face images under varying poses still remains an open research area. Face recognition across pose is the ability of a face recognition technology (FRT) to recognise face images in different viewpoints, i.e. recognition of face images that are out of the image plane. In this research work, a short literature survey of 2D techniques which are used to correct pose are discussed. The classification of these techniques is based on three categories, (1) Real-view based matching, (2) Image space pose transformation and (3) Feature space pose transformation. This paper discuss the types of databases used, approaches to correct pose, the types of features extracteden_US
dc.language.isoenen_US
dc.publisherIEEEen_US
dc.subjectFace recognitionen_US
dc.subjectPose invarianten_US
dc.subject2.0en_US
dc.title2D methods for pose invariant face recognitionen_US
dc.typePresentationen_US
dc.contributor.researchID12363626 - Helberg, Albertus Stephanus Jacobus


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record