A Nonlinear Discriminative Approach to AAM Fitting

Jason Sarigih; Roland Goecke

doi:10.1109/ICCV.2007.4409106

A Nonlinear Discriminative Approach to AAM Fitting

Jason Sarigih, Roland Goecke

Research output: A Conference proceeding or a Chapter in Book › Conference contribution › peer-review

132 Citations (Scopus)

52 Downloads (Pure)

Abstract

The Active Appearance Model (AAM) is a powerful generative method for modeling and registering deformable visual objects. Most methods for AAM fitting utilize a linear
parameter update model in an iterative framework. Despite its popularity, the scope of this approach is severely
restricted, both in fitting accuracy and capture range, due
to the simplicity of the linear update models used. In this
paper, we present an new AAM fitting formulation, which
utilizes a nonlinear update model. To motivate our approach, we compare its performance against two popular
fitting methods on two publicly available face databases, in
which this formulation boasts significant performance improvements.

Original language	English
Title of host publication	Proceedings of the Eleventh IEEE International Conference on Computer Vision ICCV2007
Place of Publication	Australia
Publisher	IEEE, Institute of Electrical and Electronics Engineers
Pages	1-8
Number of pages	8
DOIs	https://doi.org/10.1109/ICCV.2007.4409106
Publication status	Published - 2007
Externally published	Yes
Event	ICCV2007 - Rio de Janeiro, Brazil Duration: 14 Oct 2007 → 20 Oct 2007

Conference

Conference	ICCV2007
Country/Territory	Brazil
City	Rio de Janeiro
Period	14/10/07 → 20/10/07

Access to Document

10.1109/ICCV.2007.4409106

PaperFinal published version, 1.16 MB

Cite this

@inproceedings{5847460308f04f838f889d1ab0a338a7,

title = "A Nonlinear Discriminative Approach to AAM Fitting",

abstract = "The Active Appearance Model (AAM) is a powerful generative method for modeling and registering deformable visual objects. Most methods for AAM fitting utilize a linearparameter update model in an iterative framework. Despite its popularity, the scope of this approach is severelyrestricted, both in fitting accuracy and capture range, dueto the simplicity of the linear update models used. In thispaper, we present an new AAM fitting formulation, whichutilizes a nonlinear update model. To motivate our approach, we compare its performance against two popularfitting methods on two publicly available face databases, inwhich this formulation boasts significant performance improvements.",

author = "Jason Sarigih and Roland Goecke",

year = "2007",

doi = "10.1109/ICCV.2007.4409106",

language = "English",

pages = "1--8",

booktitle = "Proceedings of the Eleventh IEEE International Conference on Computer Vision ICCV2007",

publisher = "IEEE, Institute of Electrical and Electronics Engineers",

address = "United States",

note = "ICCV2007 ; Conference date: 14-10-2007 Through 20-10-2007",

}

TY - GEN

T1 - A Nonlinear Discriminative Approach to AAM Fitting

AU - Sarigih, Jason

AU - Goecke, Roland

PY - 2007

Y1 - 2007

N2 - The Active Appearance Model (AAM) is a powerful generative method for modeling and registering deformable visual objects. Most methods for AAM fitting utilize a linearparameter update model in an iterative framework. Despite its popularity, the scope of this approach is severelyrestricted, both in fitting accuracy and capture range, dueto the simplicity of the linear update models used. In thispaper, we present an new AAM fitting formulation, whichutilizes a nonlinear update model. To motivate our approach, we compare its performance against two popularfitting methods on two publicly available face databases, inwhich this formulation boasts significant performance improvements.

AB - The Active Appearance Model (AAM) is a powerful generative method for modeling and registering deformable visual objects. Most methods for AAM fitting utilize a linearparameter update model in an iterative framework. Despite its popularity, the scope of this approach is severelyrestricted, both in fitting accuracy and capture range, dueto the simplicity of the linear update models used. In thispaper, we present an new AAM fitting formulation, whichutilizes a nonlinear update model. To motivate our approach, we compare its performance against two popularfitting methods on two publicly available face databases, inwhich this formulation boasts significant performance improvements.

U2 - 10.1109/ICCV.2007.4409106

DO - 10.1109/ICCV.2007.4409106

M3 - Conference contribution

SP - 1

EP - 8

BT - Proceedings of the Eleventh IEEE International Conference on Computer Vision ICCV2007

PB - IEEE, Institute of Electrical and Electronics Engineers

CY - Australia

T2 - ICCV2007

Y2 - 14 October 2007 through 20 October 2007

ER -