User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

Independent component analysis for face authentication

Primary tabs

Havran, Carmen [UCL] Hupet, Laurent [UCL] Czyz, Jacek [UCL] Lee, John Aldo [UCL] Vandendorpe, Luc [UCL] Verleysen, Michel [UCL]

Independent component analysis (ICA) is presented as an alternative feature extraction algorithm to principal component analysis (PCA) widely used in automatic face recognition/authentication tasks. We show that the promising ICA algorithm extracts from faces features that are relevant and efficient for authentication. This leads to improved success rates and a reduced client model size over a PCA based feature extraction.

Document type Communication à un colloque (Conference Paper) – Présentation orale avec comité de sélection
Access type Accès restreint
Publication date 2002
Language Anglais
Conference "International Conference on Knowledge-Based Intelligent Information Engineering Systems (KES 2002)", Crema (Italy) (du 16/09/2002 au 18/09/2002)
Peer reviewed yes
Host document Damiani, E.; Howlett, R.J.; Jain, L.C.; Ichalkaranje, N. ; "Knowledge-Based Intelligent Information Engineering Systems and Allied Technologies"- 1207-1211 (ISBN : 1-58603280-1)
Publisher IOS press
Affiliations UCL - MD/RAIM - Département de radiologie et d'imagerie médicale
UCL - FSA/ELEC - Département d'électricité
Links
Bibliographic reference Havran, Carmen ; Hupet, Laurent ; Czyz, Jacek ; Lee, John Aldo ; Vandendorpe, Luc ; et. al. Independent component analysis for face authentication.International Conference on Knowledge-Based Intelligent Information Engineering Systems (KES 2002) (Crema (Italy), du 16/09/2002 au 18/09/2002). In: Damiani, E.; Howlett, R.J.; Jain, L.C.; Ichalkaranje, N., Knowledge-Based Intelligent Information Engineering Systems and Allied Technologies, IOS press2002, p.1207-1211
Permanent URL http://hdl.handle.net/2078.1/68039