How can we reconstruct facial image from partially occluded or low-resolution one?

Seong Whan Lee, Jeong Seon Park, Bon W. Hwang

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

This paper presents our method for reconstructing facial image from a partially occluded facial image or a low-resolution one using example-based learning. Faces are modeled by linear combinations of prototypes of shape and texture. With the shape and texture information from an input facial image, we can estimate optimal coefficients for linear combinations of prototypes of shape and texture by simple projection for least square minimization. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by reconstructing facial image from a partially occluded facial image or a low-resolution one.

Original languageEnglish
Pages (from-to)386-399
Number of pages14
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3338
Publication statusPublished - 2004 Dec 1

Fingerprint

Textures
Texture
Linear Combination
Face recognition
Least-Squares Analysis
Prototype
Learning
Face Recognition
Least Squares
Projection
Coefficient
Estimate
Facial Recognition

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)
  • Computer Science(all)
  • Theoretical Computer Science

Cite this

@article{e315474593dc471daecc9257666aa205,
title = "How can we reconstruct facial image from partially occluded or low-resolution one?",
abstract = "This paper presents our method for reconstructing facial image from a partially occluded facial image or a low-resolution one using example-based learning. Faces are modeled by linear combinations of prototypes of shape and texture. With the shape and texture information from an input facial image, we can estimate optimal coefficients for linear combinations of prototypes of shape and texture by simple projection for least square minimization. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by reconstructing facial image from a partially occluded facial image or a low-resolution one.",
author = "Lee, {Seong Whan} and Park, {Jeong Seon} and Hwang, {Bon W.}",
year = "2004",
month = "12",
day = "1",
language = "English",
volume = "3338",
pages = "386--399",
journal = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
issn = "0302-9743",
publisher = "Springer Verlag",

}

TY - JOUR

T1 - How can we reconstruct facial image from partially occluded or low-resolution one?

AU - Lee, Seong Whan

AU - Park, Jeong Seon

AU - Hwang, Bon W.

PY - 2004/12/1

Y1 - 2004/12/1

N2 - This paper presents our method for reconstructing facial image from a partially occluded facial image or a low-resolution one using example-based learning. Faces are modeled by linear combinations of prototypes of shape and texture. With the shape and texture information from an input facial image, we can estimate optimal coefficients for linear combinations of prototypes of shape and texture by simple projection for least square minimization. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by reconstructing facial image from a partially occluded facial image or a low-resolution one.

AB - This paper presents our method for reconstructing facial image from a partially occluded facial image or a low-resolution one using example-based learning. Faces are modeled by linear combinations of prototypes of shape and texture. With the shape and texture information from an input facial image, we can estimate optimal coefficients for linear combinations of prototypes of shape and texture by simple projection for least square minimization. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by reconstructing facial image from a partially occluded facial image or a low-resolution one.

UR - http://www.scopus.com/inward/record.url?scp=35048867841&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048867841&partnerID=8YFLogxK

M3 - Article

VL - 3338

SP - 386

EP - 399

JO - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

JF - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SN - 0302-9743

ER -