Posed face image synthesis using nonlinear manifold learning

Eunok Cho, Daijin Kim, Sang Youn Lee

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

This paper proposes to synthesize posed facial images from two parameters for the pose. This parameterization makes the representation, storage, and transmission of face images effective. Because variations of face images show a complicated nonlinear manifold in high-dimensional data space, we use an LLE (Locally Linear Embedding) technique for a good representation of face images. And we apply a snake model to estimate face feature values in the reduced feature space that corresponds to a specific pose parameter. Finally, a synthetic face image is obtained from an interpolation of several neighboring face images. Experimental results show that the proposed method creates an accurate and consistent synthetic face images with respect to changes of pose.

Original languageEnglish
Pages (from-to)946-954
Number of pages9
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2688
Publication statusPublished - 2003 Dec 1

Fingerprint

Manifold Learning
Parameterization
Interpolation
Face
Synthesis
Locally Linear Embedding
Snakes
High-dimensional Data
Feature Space
Two Parameters
Interpolate
Experimental Results
Estimate

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

@article{5b2d5bdd8a6742838527c7a44689a0e7,
title = "Posed face image synthesis using nonlinear manifold learning",
abstract = "This paper proposes to synthesize posed facial images from two parameters for the pose. This parameterization makes the representation, storage, and transmission of face images effective. Because variations of face images show a complicated nonlinear manifold in high-dimensional data space, we use an LLE (Locally Linear Embedding) technique for a good representation of face images. And we apply a snake model to estimate face feature values in the reduced feature space that corresponds to a specific pose parameter. Finally, a synthetic face image is obtained from an interpolation of several neighboring face images. Experimental results show that the proposed method creates an accurate and consistent synthetic face images with respect to changes of pose.",
author = "Eunok Cho and Daijin Kim and Lee, {Sang Youn}",
year = "2003",
month = "12",
day = "1",
language = "English",
volume = "2688",
pages = "946--954",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - Posed face image synthesis using nonlinear manifold learning

AU - Cho, Eunok

AU - Kim, Daijin

AU - Lee, Sang Youn

PY - 2003/12/1

Y1 - 2003/12/1

N2 - This paper proposes to synthesize posed facial images from two parameters for the pose. This parameterization makes the representation, storage, and transmission of face images effective. Because variations of face images show a complicated nonlinear manifold in high-dimensional data space, we use an LLE (Locally Linear Embedding) technique for a good representation of face images. And we apply a snake model to estimate face feature values in the reduced feature space that corresponds to a specific pose parameter. Finally, a synthetic face image is obtained from an interpolation of several neighboring face images. Experimental results show that the proposed method creates an accurate and consistent synthetic face images with respect to changes of pose.

AB - This paper proposes to synthesize posed facial images from two parameters for the pose. This parameterization makes the representation, storage, and transmission of face images effective. Because variations of face images show a complicated nonlinear manifold in high-dimensional data space, we use an LLE (Locally Linear Embedding) technique for a good representation of face images. And we apply a snake model to estimate face feature values in the reduced feature space that corresponds to a specific pose parameter. Finally, a synthetic face image is obtained from an interpolation of several neighboring face images. Experimental results show that the proposed method creates an accurate and consistent synthetic face images with respect to changes of pose.

UR - http://www.scopus.com/inward/record.url?scp=33745620410&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745620410&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33745620410

VL - 2688

SP - 946

EP - 954

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

ER -