Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models

Moritz Kaiser; Nicolas Lehment; Gerhard Rigoll

doi:10.1109/ICIP.2011.6116705

Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models

Moritz Kaiser, Nicolas Lehment, Gerhard Rigoll

Chair of Human-Machine Communication

Technical University of Munich

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

Abstract

In this contribution a novel method to compute dense point-to-point correspondences between 3D faces is presented. The faces are aligned in 3D space with a Generalized Procrustes Analysis and subsequently mapped into 2D space. To compute a correspondence flow between two faces an energy function is minimized which is based on the following assumptions: smoothness of the flow, mapping of landmarks on their counterparts, and texture and depth consistency. Based on these correspondences, the 3D faces are resampled, so that each face is represented by the same amount of 3D points and for any point there is a corresponding point in all other faces. The accuracy of the point-to-point correspondences is demonstrated on the basis of two applications, namely facial texture mapping and the construction of 3D Morphable Models.

Original language	English
Title of host publication	ICIP 2011
Subtitle of host publication	2011 18th IEEE International Conference on Image Processing
Pages	901-904
Number of pages	4
DOIs	https://doi.org/10.1109/ICIP.2011.6116705
State	Published - 2011
Event	2011 18th IEEE International Conference on Image Processing, ICIP 2011 - Brussels, Belgium Duration: 11 Sep 2011 → 14 Sep 2011

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
ISSN (Print)	1522-4880

Conference

Conference	2011 18th IEEE International Conference on Image Processing, ICIP 2011
Country/Territory	Belgium
City	Brussels
Period	11/09/11 → 14/09/11

Keywords

3D surface registration
correspondence estimation
face image processing
face synthesis

Access to Document

10.1109/ICIP.2011.6116705

Cite this

Kaiser, M., Lehment, N., & Rigoll, G. (2011). Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models. In ICIP 2011: 2011 18th IEEE International Conference on Image Processing (pp. 901-904). Article 6116705 (Proceedings - International Conference on Image Processing, ICIP). https://doi.org/10.1109/ICIP.2011.6116705

@inproceedings{ea53bb6024c64aa8864f72c02d733b11,

title = "Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models",

abstract = "In this contribution a novel method to compute dense point-to-point correspondences between 3D faces is presented. The faces are aligned in 3D space with a Generalized Procrustes Analysis and subsequently mapped into 2D space. To compute a correspondence flow between two faces an energy function is minimized which is based on the following assumptions: smoothness of the flow, mapping of landmarks on their counterparts, and texture and depth consistency. Based on these correspondences, the 3D faces are resampled, so that each face is represented by the same amount of 3D points and for any point there is a corresponding point in all other faces. The accuracy of the point-to-point correspondences is demonstrated on the basis of two applications, namely facial texture mapping and the construction of 3D Morphable Models.",

keywords = "3D surface registration, correspondence estimation, face image processing, face synthesis",

author = "Moritz Kaiser and Nicolas Lehment and Gerhard Rigoll",

year = "2011",

doi = "10.1109/ICIP.2011.6116705",

language = "English",

isbn = "9781457713033",

series = "Proceedings - International Conference on Image Processing, ICIP",

pages = "901--904",

booktitle = "ICIP 2011",

note = "2011 18th IEEE International Conference on Image Processing, ICIP 2011 ; Conference date: 11-09-2011 Through 14-09-2011",

}

Kaiser, M, Lehment, N & Rigoll, G 2011, Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models. in ICIP 2011: 2011 18th IEEE International Conference on Image Processing., 6116705, Proceedings - International Conference on Image Processing, ICIP, pp. 901-904, 2011 18th IEEE International Conference on Image Processing, ICIP 2011, Brussels, Belgium, 11/09/11. https://doi.org/10.1109/ICIP.2011.6116705

Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models. / Kaiser, Moritz; Lehment, Nicolas; Rigoll, Gerhard.
ICIP 2011: 2011 18th IEEE International Conference on Image Processing. 2011. p. 901-904 6116705 (Proceedings - International Conference on Image Processing, ICIP).

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

TY - GEN

T1 - Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models

AU - Kaiser, Moritz

AU - Lehment, Nicolas

AU - Rigoll, Gerhard

PY - 2011

Y1 - 2011

N2 - In this contribution a novel method to compute dense point-to-point correspondences between 3D faces is presented. The faces are aligned in 3D space with a Generalized Procrustes Analysis and subsequently mapped into 2D space. To compute a correspondence flow between two faces an energy function is minimized which is based on the following assumptions: smoothness of the flow, mapping of landmarks on their counterparts, and texture and depth consistency. Based on these correspondences, the 3D faces are resampled, so that each face is represented by the same amount of 3D points and for any point there is a corresponding point in all other faces. The accuracy of the point-to-point correspondences is demonstrated on the basis of two applications, namely facial texture mapping and the construction of 3D Morphable Models.

AB - In this contribution a novel method to compute dense point-to-point correspondences between 3D faces is presented. The faces are aligned in 3D space with a Generalized Procrustes Analysis and subsequently mapped into 2D space. To compute a correspondence flow between two faces an energy function is minimized which is based on the following assumptions: smoothness of the flow, mapping of landmarks on their counterparts, and texture and depth consistency. Based on these correspondences, the 3D faces are resampled, so that each face is represented by the same amount of 3D points and for any point there is a corresponding point in all other faces. The accuracy of the point-to-point correspondences is demonstrated on the basis of two applications, namely facial texture mapping and the construction of 3D Morphable Models.

KW - 3D surface registration

KW - correspondence estimation

KW - face image processing

KW - face synthesis

UR - http://www.scopus.com/inward/record.url?scp=84856237796&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2011.6116705

DO - 10.1109/ICIP.2011.6116705

M3 - Conference contribution

AN - SCOPUS:84856237796

SN - 9781457713033

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 901

EP - 904

BT - ICIP 2011

T2 - 2011 18th IEEE International Conference on Image Processing, ICIP 2011

Y2 - 11 September 2011 through 14 September 2011

ER -

Dense point-to-point correspondences between 3D faces with large variations for constructing 3D morphable models

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this