Face recognition under pose variations using shape-adapted texture features

Publikationstyp
Conference Paper
Publikationsdatum
2010-12-03
Sprache
English
Author
Gernoth, Thorsten 
Gooßen, André 
Grigat, Rolf-Rainer 
Institut
Bildverarbeitungssysteme E-2 
TORE-URI
http://hdl.handle.net/11420/10036
First published in
Proceedings - International Conference on Image Processing, ICIP 
Start Page
4525
End Page
4528
Article Number
5651335
Citation
Proceedings - International Conference on Image Processing, ICIP: 5651335, 4525-4528 (2010-12)
Contribution to Conference
17th IEEE International Conference on Image Processing (ICIP 2010) 
Publisher DOI
10.1109/ICIP.2010.5651335
Scopus ID
2-s2.0-78651069610
Publisher
IEEE
The complexity in face recognition emerges from the variability of the appearance of a human face. While the identity is preserved, the appearance of a face may change due to factors such as illumination, pose or facial expression. To recognize a person independent of pose, we want to separate shape from texture information. We concentrate on the texture part in this work. We first fit an active appearance model to a given facial image. The shape information is used to transform the face into a shape-free representation. We decompose the transformed face into local regions and extract texture features from these not necessarily rectangular regions using a shape-adapted discrete cosine transform. The texture features we use for face recognition are independent of pose and shape of the face. We show that these features contain sufficient discriminative information to recognize persons across changes in pose. © 2010 IEEE.
Schlagworte
Active appearance model
Discrete cosine transforms
Face recognition
DDC Class
004: Informatik