Face Alignment Using Boosting and Evolutionary Search

Hua Zhang; Duanduan Liu; Mannes Poel; Antinus Nijholt

doi:10.1007/978-3-642-12304-7_11

Face Alignment Using Boosting and Evolutionary Search

Hua Zhang, Duanduan Liu, Mannes Poel, Antinus Nijholt

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Citation (Scopus)

222 Downloads (Pure)

Abstract

In this paper, we present a face alignment approach using granular features, boosting, and an evolutionary search algorithm. Active Appearance Models (AAM) integrate a shape-texture-combined morphable face model into an efficient fitting strategy, then Boosting Appearance Models (BAM) consider the face alignment problem as a process of maximizing the response from a boosting classifier. Enlightened by AAM and BAM, we present a framework which implements improved boosting classifiers based on more discriminative features and exhaustive search strategies. In this paper, we utilize granular features to replace the conventional rectangular Haar-like features, to improve discriminability, computational efficiency, and a larger search space. At the same time, we adopt the evolutionary search process to solve the deficiency of searching in the large feature space. Finally, we test our approach on a series of challenging data sets, to show the accuracy and efficiency on versatile face images.

Original language	Undefined
Title of host publication	Ninth Asian Conference on Computer Vision (ACCV 2009). Part II
Editors	H. Zha, R.-I. Taniguchi, S. Maybank
Place of Publication	Berlin
Publisher	Springer
Pages	110-119
Number of pages	10
ISBN (Print)	978-3-642-12303-0
DOIs	https://doi.org/10.1007/978-3-642-12304-7_11
Publication status	Published - 25 Apr 2010
Event	9th Asian Conference on Computer Vision, ACCV 2009 - Peking University, Xi'an, China Duration: 23 Sept 2009 → 27 Sept 2009 Conference number: 9

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer Verlag
Volume	5995

Conference

Conference	9th Asian Conference on Computer Vision, ACCV 2009
Abbreviated title	ACCV
Country/Territory	China
City	Xi'an
Period	23/09/09 → 27/09/09

Keywords

METIS-270691
IR-71163
evolutionary search
Face alignment
EWI-16061
granular features
boosting appearance models
HMI-MI: MULTIMODAL INTERACTIONS
EC Grant Agreement nr.: FP6/033812

Access to Document

10.1007/978-3-642-12304-7_11

accv2009

http://eprints.eemcs.utwente.nl/secure2/16061/01/accv2009.pdf

Cite this

@inproceedings{a7f9fa59e9e54db793edee1ad5d76522,

title = "Face Alignment Using Boosting and Evolutionary Search",

abstract = "In this paper, we present a face alignment approach using granular features, boosting, and an evolutionary search algorithm. Active Appearance Models (AAM) integrate a shape-texture-combined morphable face model into an efficient fitting strategy, then Boosting Appearance Models (BAM) consider the face alignment problem as a process of maximizing the response from a boosting classifier. Enlightened by AAM and BAM, we present a framework which implements improved boosting classifiers based on more discriminative features and exhaustive search strategies. In this paper, we utilize granular features to replace the conventional rectangular Haar-like features, to improve discriminability, computational efficiency, and a larger search space. At the same time, we adopt the evolutionary search process to solve the deficiency of searching in the large feature space. Finally, we test our approach on a series of challenging data sets, to show the accuracy and efficiency on versatile face images.",

keywords = "METIS-270691, IR-71163, evolutionary search, Face alignment, EWI-16061, granular features, boosting appearance models, HMI-MI: MULTIMODAL INTERACTIONS, EC Grant Agreement nr.: FP6/033812",

author = "Hua Zhang and Duanduan Liu and Mannes Poel and Antinus Nijholt",

note = "10.1007/978-3-642-12304-7_11 ; 9th Asian Conference on Computer Vision, ACCV 2009 ; Conference date: 23-09-2009 Through 27-09-2009",

year = "2010",

month = apr,

day = "25",

doi = "10.1007/978-3-642-12304-7_11",

language = "Undefined",

isbn = "978-3-642-12303-0",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "110--119",

editor = "H. Zha and R.-I. Taniguchi and S. Maybank",

booktitle = "Ninth Asian Conference on Computer Vision (ACCV 2009). Part II",

address = "Germany",

}

Zhang, H, Liu, D, Poel, M & Nijholt, A 2010, Face Alignment Using Boosting and Evolutionary Search. in H Zha, R-I Taniguchi & S Maybank (eds), Ninth Asian Conference on Computer Vision (ACCV 2009). Part II. Lecture Notes in Computer Science, vol. 5995, Springer, Berlin, pp. 110-119, 9th Asian Conference on Computer Vision, ACCV 2009, Xi'an, China, 23/09/09. https://doi.org/10.1007/978-3-642-12304-7_11

Face Alignment Using Boosting and Evolutionary Search. / Zhang, Hua; Liu, Duanduan; Poel, Mannes et al.
Ninth Asian Conference on Computer Vision (ACCV 2009). Part II. ed. / H. Zha; R.-I. Taniguchi; S. Maybank. Berlin: Springer, 2010. p. 110-119 (Lecture Notes in Computer Science; Vol. 5995).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Face Alignment Using Boosting and Evolutionary Search

AU - Zhang, Hua

AU - Liu, Duanduan

AU - Poel, Mannes

AU - Nijholt, Antinus

N1 - Conference code: 9

PY - 2010/4/25

Y1 - 2010/4/25

N2 - In this paper, we present a face alignment approach using granular features, boosting, and an evolutionary search algorithm. Active Appearance Models (AAM) integrate a shape-texture-combined morphable face model into an efficient fitting strategy, then Boosting Appearance Models (BAM) consider the face alignment problem as a process of maximizing the response from a boosting classifier. Enlightened by AAM and BAM, we present a framework which implements improved boosting classifiers based on more discriminative features and exhaustive search strategies. In this paper, we utilize granular features to replace the conventional rectangular Haar-like features, to improve discriminability, computational efficiency, and a larger search space. At the same time, we adopt the evolutionary search process to solve the deficiency of searching in the large feature space. Finally, we test our approach on a series of challenging data sets, to show the accuracy and efficiency on versatile face images.

AB - In this paper, we present a face alignment approach using granular features, boosting, and an evolutionary search algorithm. Active Appearance Models (AAM) integrate a shape-texture-combined morphable face model into an efficient fitting strategy, then Boosting Appearance Models (BAM) consider the face alignment problem as a process of maximizing the response from a boosting classifier. Enlightened by AAM and BAM, we present a framework which implements improved boosting classifiers based on more discriminative features and exhaustive search strategies. In this paper, we utilize granular features to replace the conventional rectangular Haar-like features, to improve discriminability, computational efficiency, and a larger search space. At the same time, we adopt the evolutionary search process to solve the deficiency of searching in the large feature space. Finally, we test our approach on a series of challenging data sets, to show the accuracy and efficiency on versatile face images.

KW - METIS-270691

KW - IR-71163

KW - evolutionary search

KW - Face alignment

KW - EWI-16061

KW - granular features

KW - boosting appearance models

KW - HMI-MI: MULTIMODAL INTERACTIONS

KW - EC Grant Agreement nr.: FP6/033812

U2 - 10.1007/978-3-642-12304-7_11

DO - 10.1007/978-3-642-12304-7_11

M3 - Conference contribution

SN - 978-3-642-12303-0

T3 - Lecture Notes in Computer Science

SP - 110

EP - 119

BT - Ninth Asian Conference on Computer Vision (ACCV 2009). Part II

A2 - Zha, H.

A2 - Taniguchi, R.-I.

A2 - Maybank, S.

PB - Springer

CY - Berlin

T2 - 9th Asian Conference on Computer Vision, ACCV 2009

Y2 - 23 September 2009 through 27 September 2009

ER -