DocumentCode :
2371064
Title :
Evolution Strategies for Matching Active Appearance Models to Human Faces
Author :
Lui, Yui Man ; Beveridge, J. Ross ; Howe, Adele E. ; Whitley, L. Darrell
Author_Institution :
Colorado State Univ., Fort Collins
fYear :
2007
fDate :
27-29 Sept. 2007
Firstpage :
1
Lastpage :
7
Abstract :
Face registration is a challenging problem due in part to the non-rigid nature of human faces. Active Appearance Models (AAMs) have been proposed as a useful technique for face registration in part because they can account for changes in shape. Fitting of AAMs to imagery is typically done using the Gauss-Newton method; however this approach is known to fail when either the initial shape estimate is to far off and/or the appearance model fails to direct search toward a good match. In this paper, we employ Evolution Strategies (ES) to search for a near optimal fit, i.e. set of model parameters, that relate an AAM to a novel face image. In addition, we dramatically reduce the dimensionality of the search problem by analytically determining the optimal texture parameters associated with any given shape. Experimental results reveal that the proposed approach outperforms the standard AAM parameter estimation technique and some of its variants. Further, the difference is statistically significant. Finally, some basic limitations of AAMs are identified using fitness landscape analysis.
Keywords :
Newton method; face recognition; image matching; image registration; image texture; search problems; Gauss-Newton method; active appearance model matching; evolution strategy; fitness landscape analysis; human face registration; optimal texture parameter estimation; search problem; Active appearance model; Active shape model; Face; Humans; Image texture analysis; Least squares methods; Newton method; Parameter estimation; Recursive estimation; Search problems;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Biometrics: Theory, Applications, and Systems, 2007. BTAS 2007. First IEEE International Conference on
Conference_Location :
Crystal City, VA
Print_ISBN :
978-1-4244-1596-0
Electronic_ISBN :
978-1-4244-1597-7
Type :
conf
DOI :
10.1109/BTAS.2007.4401922
Filename :
4401922
Link To Document :
بازگشت