Application of 3D Morphable Models to faces in video images

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

31 Downloads (Pure)

Abstract

The 3D Morphable Face Model (3DMM) has been used for over a decade for creating 3D models from single images of faces. This model is based on a PCA model of the 3D shape and texture generated from a limited number of 3D scans. The goal of fitting a 3DMM to an image is to find the model coefficients, the lighting and other imaging variables from which we can remodel that image as accurately as possible. These coefficients can without further processing be used in verification and recognition experiments. In this paper, we investigate the potential benefits of using multiple images from the same person to fit a 3DMM. Lighting and imaging variables can differ from image to image, but the PCA coefficients will remain the same, assuming no change in expression between the images. We expect using multiple images could result in a more accurate fit.
Original languageEnglish
Title of host publication32nd WIC Symposium on Information Theory in the Benelux
EditorsOlivier van den Biggelaar
Place of PublicationBrussels
PublisherWerkgemeenschap voor Informatie- en Communicatietheorie (WIC)
Pages34-41
Number of pages8
ISBN (Print)978-9-08172-190-5
Publication statusPublished - May 2011
Event32nd WIC Symposium on Information Theory in the Benelux 2011 - Brussels, Belgium
Duration: 10 May 201111 May 2011
Conference number: 32

Conference

Conference32nd WIC Symposium on Information Theory in the Benelux 2011
CountryBelgium
CityBrussels
Period10/05/1111/05/11

Fingerprint

Imaging techniques
Textures
Lighting
Processing
Experiments

Keywords

  • METIS-277640
  • Morphable Models
  • Image Processing
  • EWI-20172
  • 3D reconstruction
  • IR-77273
  • SCS-Safety
  • Biometrics

Cite this

van Rootseler, R. T. A., Spreeuwers, L. J., & Veldhuis, R. N. J. (2011). Application of 3D Morphable Models to faces in video images. In O. van den Biggelaar (Ed.), 32nd WIC Symposium on Information Theory in the Benelux (pp. 34-41). Brussels: Werkgemeenschap voor Informatie- en Communicatietheorie (WIC).
van Rootseler, R.T.A. ; Spreeuwers, Lieuwe Jan ; Veldhuis, Raymond N.J. / Application of 3D Morphable Models to faces in video images. 32nd WIC Symposium on Information Theory in the Benelux. editor / Olivier van den Biggelaar. Brussels : Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2011. pp. 34-41
@inproceedings{f06dd3c9f4b943428191677db3014d2e,
title = "Application of 3D Morphable Models to faces in video images",
abstract = "The 3D Morphable Face Model (3DMM) has been used for over a decade for creating 3D models from single images of faces. This model is based on a PCA model of the 3D shape and texture generated from a limited number of 3D scans. The goal of fitting a 3DMM to an image is to find the model coefficients, the lighting and other imaging variables from which we can remodel that image as accurately as possible. These coefficients can without further processing be used in verification and recognition experiments. In this paper, we investigate the potential benefits of using multiple images from the same person to fit a 3DMM. Lighting and imaging variables can differ from image to image, but the PCA coefficients will remain the same, assuming no change in expression between the images. We expect using multiple images could result in a more accurate fit.",
keywords = "METIS-277640, Morphable Models, Image Processing, EWI-20172, 3D reconstruction, IR-77273, SCS-Safety, Biometrics",
author = "{van Rootseler}, R.T.A. and Spreeuwers, {Lieuwe Jan} and Veldhuis, {Raymond N.J.}",
note = "http://opera.ulb.ac.be/wicsp2011/free/WICSP2011_Proceedings.pdf",
year = "2011",
month = "5",
language = "English",
isbn = "978-9-08172-190-5",
pages = "34--41",
editor = "{van den Biggelaar}, Olivier",
booktitle = "32nd WIC Symposium on Information Theory in the Benelux",
publisher = "Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)",
address = "Netherlands",

}

van Rootseler, RTA, Spreeuwers, LJ & Veldhuis, RNJ 2011, Application of 3D Morphable Models to faces in video images. in O van den Biggelaar (ed.), 32nd WIC Symposium on Information Theory in the Benelux. Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), Brussels, pp. 34-41, 32nd WIC Symposium on Information Theory in the Benelux 2011, Brussels, Belgium, 10/05/11.

Application of 3D Morphable Models to faces in video images. / van Rootseler, R.T.A.; Spreeuwers, Lieuwe Jan; Veldhuis, Raymond N.J.

32nd WIC Symposium on Information Theory in the Benelux. ed. / Olivier van den Biggelaar. Brussels : Werkgemeenschap voor Informatie- en Communicatietheorie (WIC), 2011. p. 34-41.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

TY - GEN

T1 - Application of 3D Morphable Models to faces in video images

AU - van Rootseler, R.T.A.

AU - Spreeuwers, Lieuwe Jan

AU - Veldhuis, Raymond N.J.

N1 - http://opera.ulb.ac.be/wicsp2011/free/WICSP2011_Proceedings.pdf

PY - 2011/5

Y1 - 2011/5

N2 - The 3D Morphable Face Model (3DMM) has been used for over a decade for creating 3D models from single images of faces. This model is based on a PCA model of the 3D shape and texture generated from a limited number of 3D scans. The goal of fitting a 3DMM to an image is to find the model coefficients, the lighting and other imaging variables from which we can remodel that image as accurately as possible. These coefficients can without further processing be used in verification and recognition experiments. In this paper, we investigate the potential benefits of using multiple images from the same person to fit a 3DMM. Lighting and imaging variables can differ from image to image, but the PCA coefficients will remain the same, assuming no change in expression between the images. We expect using multiple images could result in a more accurate fit.

AB - The 3D Morphable Face Model (3DMM) has been used for over a decade for creating 3D models from single images of faces. This model is based on a PCA model of the 3D shape and texture generated from a limited number of 3D scans. The goal of fitting a 3DMM to an image is to find the model coefficients, the lighting and other imaging variables from which we can remodel that image as accurately as possible. These coefficients can without further processing be used in verification and recognition experiments. In this paper, we investigate the potential benefits of using multiple images from the same person to fit a 3DMM. Lighting and imaging variables can differ from image to image, but the PCA coefficients will remain the same, assuming no change in expression between the images. We expect using multiple images could result in a more accurate fit.

KW - METIS-277640

KW - Morphable Models

KW - Image Processing

KW - EWI-20172

KW - 3D reconstruction

KW - IR-77273

KW - SCS-Safety

KW - Biometrics

M3 - Conference contribution

SN - 978-9-08172-190-5

SP - 34

EP - 41

BT - 32nd WIC Symposium on Information Theory in the Benelux

A2 - van den Biggelaar, Olivier

PB - Werkgemeenschap voor Informatie- en Communicatietheorie (WIC)

CY - Brussels

ER -

van Rootseler RTA, Spreeuwers LJ, Veldhuis RNJ. Application of 3D Morphable Models to faces in video images. In van den Biggelaar O, editor, 32nd WIC Symposium on Information Theory in the Benelux. Brussels: Werkgemeenschap voor Informatie- en Communicatietheorie (WIC). 2011. p. 34-41