Efficient Robust Active Appearance Model Fitting

Markus Storer, Peter Roth, Martin Urschler, Horst Bischof, Josef Alois Birchbauer

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The Active Appearance Model (AAM) is a widely used approach for model based vision showing excellent results. But one major drawback is that the method is not robust against occlusions. Thus, if parts of the image are occluded the method converges to local minima and the obtained results are unreliable. To overcome this problem we propose a robust AAM fitting strategy. The main idea is to apply a robust PCA model to reconstruct the missing feature information and to use the thus obtained image as input for the standard AAM fitting process. Since existing methods for robust PCA reconstruction are computationally too expensive for real-time processing we applied a more efficient method: Fast-Robust PCA (FR-PCA). In fact, by using our FR-PCA the computational effort is drastically reduced. Moreover, more accurate reconstructions are obtained. In the experiments, we evaluated both, the FR-PCA model on the publicly available ALOI database and the whole robust AAM fitting chain on facial images. The results clearly show the benefits of our approach in terms of accuracy and speed when processing disturbed data (i.e., images containing occlusions).
LanguageEnglish
Title of host publicationComputer Vision, Imaging and Computer Graphics. Theory and Applications
Subtitle of host publicationInternational Joint Conference, VISIGRAPP 2009, Lisboa, Portugal, February 5-8, 2009. Revised Selected Papers
EditorsAlpesh K. Ranchordas, Joao M. Pereira, Helder J. Araujo, Joao M. R. S. Tavares
Place of PublicationBerlin Heidelberg
PublisherSpringer
Pages229-241
Volume68
ISBN (Electronic)978-3-642-11840-1
ISBN (Print)978-3-642-11839-5
DOIs
StatusPublished - 2010

Publication series

NameCommunications in Computer and Information Science

Fingerprint

Processing
Experiments

Fields of Expertise

  • Information, Communication & Computing

Cite this

Storer, M., Roth, P., Urschler, M., Bischof, H., & Birchbauer, J. A. (2010). Efficient Robust Active Appearance Model Fitting. In A. K. Ranchordas, J. M. Pereira, H. J. Araujo, & J. M. R. S. Tavares (Eds.), Computer Vision, Imaging and Computer Graphics. Theory and Applications: International Joint Conference, VISIGRAPP 2009, Lisboa, Portugal, February 5-8, 2009. Revised Selected Papers (Vol. 68, pp. 229-241). (Communications in Computer and Information Science). Berlin Heidelberg: Springer. DOI: 10.1007/978-3-642-11840-1_17

Efficient Robust Active Appearance Model Fitting. / Storer, Markus; Roth, Peter; Urschler, Martin; Bischof, Horst; Birchbauer, Josef Alois.

Computer Vision, Imaging and Computer Graphics. Theory and Applications: International Joint Conference, VISIGRAPP 2009, Lisboa, Portugal, February 5-8, 2009. Revised Selected Papers. ed. / Alpesh K. Ranchordas; Joao M. Pereira; Helder J. Araujo; Joao M. R. S. Tavares. Vol. 68 Berlin Heidelberg : Springer, 2010. p. 229-241 (Communications in Computer and Information Science).

Research output: Chapter in Book/Report/Conference proceedingChapter

Storer, M, Roth, P, Urschler, M, Bischof, H & Birchbauer, JA 2010, Efficient Robust Active Appearance Model Fitting. in AK Ranchordas, JM Pereira, HJ Araujo & JMRS Tavares (eds), Computer Vision, Imaging and Computer Graphics. Theory and Applications: International Joint Conference, VISIGRAPP 2009, Lisboa, Portugal, February 5-8, 2009. Revised Selected Papers. vol. 68, Communications in Computer and Information Science, Springer, Berlin Heidelberg, pp. 229-241. DOI: 10.1007/978-3-642-11840-1_17
Storer M, Roth P, Urschler M, Bischof H, Birchbauer JA. Efficient Robust Active Appearance Model Fitting. In Ranchordas AK, Pereira JM, Araujo HJ, Tavares JMRS, editors, Computer Vision, Imaging and Computer Graphics. Theory and Applications: International Joint Conference, VISIGRAPP 2009, Lisboa, Portugal, February 5-8, 2009. Revised Selected Papers. Vol. 68. Berlin Heidelberg: Springer. 2010. p. 229-241. (Communications in Computer and Information Science). Available from, DOI: 10.1007/978-3-642-11840-1_17
Storer, Markus ; Roth, Peter ; Urschler, Martin ; Bischof, Horst ; Birchbauer, Josef Alois. / Efficient Robust Active Appearance Model Fitting. Computer Vision, Imaging and Computer Graphics. Theory and Applications: International Joint Conference, VISIGRAPP 2009, Lisboa, Portugal, February 5-8, 2009. Revised Selected Papers. editor / Alpesh K. Ranchordas ; Joao M. Pereira ; Helder J. Araujo ; Joao M. R. S. Tavares. Vol. 68 Berlin Heidelberg : Springer, 2010. pp. 229-241 (Communications in Computer and Information Science).
@inbook{7fe8fea871944b14b198d42700c904ae,
title = "Efficient Robust Active Appearance Model Fitting",
abstract = "The Active Appearance Model (AAM) is a widely used approach for model based vision showing excellent results. But one major drawback is that the method is not robust against occlusions. Thus, if parts of the image are occluded the method converges to local minima and the obtained results are unreliable. To overcome this problem we propose a robust AAM fitting strategy. The main idea is to apply a robust PCA model to reconstruct the missing feature information and to use the thus obtained image as input for the standard AAM fitting process. Since existing methods for robust PCA reconstruction are computationally too expensive for real-time processing we applied a more efficient method: Fast-Robust PCA (FR-PCA). In fact, by using our FR-PCA the computational effort is drastically reduced. Moreover, more accurate reconstructions are obtained. In the experiments, we evaluated both, the FR-PCA model on the publicly available ALOI database and the whole robust AAM fitting chain on facial images. The results clearly show the benefits of our approach in terms of accuracy and speed when processing disturbed data (i.e., images containing occlusions).",
author = "Markus Storer and Peter Roth and Martin Urschler and Horst Bischof and Birchbauer, {Josef Alois}",
note = "VISIGRAPP 2009, Lisboa",
year = "2010",
doi = "10.1007/978-3-642-11840-1_17",
language = "English",
isbn = "978-3-642-11839-5",
volume = "68",
series = "Communications in Computer and Information Science",
publisher = "Springer",
pages = "229--241",
editor = "Ranchordas, {Alpesh K.} and Pereira, {Joao M.} and Araujo, {Helder J.} and Tavares, {Joao M. R. S.}",
booktitle = "Computer Vision, Imaging and Computer Graphics. Theory and Applications",

}

TY - CHAP

T1 - Efficient Robust Active Appearance Model Fitting

AU - Storer,Markus

AU - Roth,Peter

AU - Urschler,Martin

AU - Bischof,Horst

AU - Birchbauer,Josef Alois

N1 - VISIGRAPP 2009, Lisboa

PY - 2010

Y1 - 2010

N2 - The Active Appearance Model (AAM) is a widely used approach for model based vision showing excellent results. But one major drawback is that the method is not robust against occlusions. Thus, if parts of the image are occluded the method converges to local minima and the obtained results are unreliable. To overcome this problem we propose a robust AAM fitting strategy. The main idea is to apply a robust PCA model to reconstruct the missing feature information and to use the thus obtained image as input for the standard AAM fitting process. Since existing methods for robust PCA reconstruction are computationally too expensive for real-time processing we applied a more efficient method: Fast-Robust PCA (FR-PCA). In fact, by using our FR-PCA the computational effort is drastically reduced. Moreover, more accurate reconstructions are obtained. In the experiments, we evaluated both, the FR-PCA model on the publicly available ALOI database and the whole robust AAM fitting chain on facial images. The results clearly show the benefits of our approach in terms of accuracy and speed when processing disturbed data (i.e., images containing occlusions).

AB - The Active Appearance Model (AAM) is a widely used approach for model based vision showing excellent results. But one major drawback is that the method is not robust against occlusions. Thus, if parts of the image are occluded the method converges to local minima and the obtained results are unreliable. To overcome this problem we propose a robust AAM fitting strategy. The main idea is to apply a robust PCA model to reconstruct the missing feature information and to use the thus obtained image as input for the standard AAM fitting process. Since existing methods for robust PCA reconstruction are computationally too expensive for real-time processing we applied a more efficient method: Fast-Robust PCA (FR-PCA). In fact, by using our FR-PCA the computational effort is drastically reduced. Moreover, more accurate reconstructions are obtained. In the experiments, we evaluated both, the FR-PCA model on the publicly available ALOI database and the whole robust AAM fitting chain on facial images. The results clearly show the benefits of our approach in terms of accuracy and speed when processing disturbed data (i.e., images containing occlusions).

U2 - 10.1007/978-3-642-11840-1_17

DO - 10.1007/978-3-642-11840-1_17

M3 - Chapter

SN - 978-3-642-11839-5

VL - 68

T3 - Communications in Computer and Information Science

SP - 229

EP - 241

BT - Computer Vision, Imaging and Computer Graphics. Theory and Applications

PB - Springer

CY - Berlin Heidelberg

ER -