Skip to main content

Research Repository

Advanced Search

Face recognition in 2D and 2.5D using ridgelets and photometric stereo

Kautkar, Satyajit N.; Atkinson, Gary A.; Smith, Melvyn L.; Kautkar, Satyajit; Atkinson, Gary; Smith, Melvyn

Face recognition in 2D and 2.5D using ridgelets and photometric stereo Thumbnail


Authors

Satyajit N. Kautkar

Gary A. Atkinson

Melvyn L. Smith

Satyajit Kautkar

Profile Image

Melvyn Smith Melvyn.Smith@uwe.ac.uk
Research Centre Director Vision Lab/Prof



Abstract

A new technique for face recognition - Ridgefaces - is presented. The method combines the well-known Fisherface method with the ridgelet transform and high-speed Photometric Stereo (PS). The paper first derives ridgelet projections for 2D/2.5D face images before the Fisherface approach is used to reduce the dimensionality and increase the spread of the resulting feature vectors. The ridgelet transform is attractive because it is efficient at extracting highly discriminating low-frequency directional features. Best recognition is obtained when Ridgefaces is performed on surface normals acquired from PS, although good results are also found using standard 2D images and PS-derived albedo maps. © 2012 Elsevier Ltd. All rights reserved.

Citation

Smith, M. L., Atkinson, G. A., Kautkar, S. N., Kautkar, S., Atkinson, G., & Smith, M. (2012). Face recognition in 2D and 2.5D using ridgelets and photometric stereo. Pattern Recognition, 45(9), 3317-3327. https://doi.org/10.1016/j.patcog.2012.03.007

Journal Article Type Article
Publication Date Jan 1, 2012
Deposit Date Jun 18, 2012
Publicly Available Date Feb 10, 2016
Journal Pattern Recognition
Print ISSN 0031-3203
Publisher Elsevier
Peer Reviewed Peer Reviewed
Volume 45
Issue 9
Pages 3317-3327
DOI https://doi.org/10.1016/j.patcog.2012.03.007
Keywords face recognition, ridgelet transform, photometric stereo, dimensionality reduction
Public URL https://uwe-repository.worktribe.com/output/944433
Publisher URL http://dx.doi.org/10.1016/j.patcog.2012.03.007

Files







You might also like



Downloadable Citations