A Robust Face Recognition Algorithm for Real-World Applications
Abstract:
The proposed face recognition algorithm utilizes representation of local facial regions with the DCT. The local representation provides robustness against appearance variations in local regions caused by partial face occlusion or facial expression, whereas utilizing the frequency information provides robustness against changes in illumination. The algorithm also bypasses the facial feature localization step and formulates face alignment as an optimization problem in the classification stage.
| Zugehörige Institution(en) am KIT | Institut für Theoretische Informatik (ITI) |
| Publikationstyp | Hochschulschrift |
| Publikationsjahr | 2009 |
| Sprache | Englisch |
| Identifikator | urn:nbn:de:swb:90-149997 KITopen-ID: 1000014999 |
| Verlag | Universität Karlsruhe (TH) |
| Art der Arbeit | Dissertation |
| Fakultät | Fakultät für Informatik (INFORMATIK) |
| Institut | Institut für Theoretische Informatik (ITI) |
| Prüfungsdaten | 02.02.2009 |
| Schlagwörter | robust face recognition, discrete cosine transform, local appearance modeling |
| Referent/Betreuer | Waibel, A. |