Reconstruction of high-resolution facial image using recursive error back-projection

Jeong Seon Park, Seong Whan Lee

Research output: Contribution to journalArticle

Abstract

This paper proposes a new method of reconstructing high-resolution facial image from a low-resolution facial image using a recursive error back-projection of example-based learning. A face is represented by a linear combination of prototypes of shape and texture. With the shape and texture information about the pixels in a given low-resolution facial image, we can estimate optimal coefficients for a linear combination of prototypes of shape and those of texture by solving least square minimization. Then high-resolution facial image can be reconstructed by using the optimal coefficients for linear combination of the high-resolution prototypes. Moreover recursive error back-projection is applied to improve the accuracy of high-resolution reconstruction. An error back-projection is composed of estimation, simulation, and error compensation. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at visual surveillance systems.

Original languageEnglish
Pages (from-to)59-66
Number of pages8
JournalLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3072
Publication statusPublished - 2004 Dec 1

Fingerprint

High Resolution
Textures
Projection
Linear Combination
Texture
Prototype
Error compensation
Face recognition
Visual Surveillance
Error Compensation
Least-Squares Analysis
Pixels
Coefficient
Face Recognition
Learning
Least Squares
Pixel
Face
Estimate
Simulation

ASJC Scopus subject areas

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

Cite this

@article{0275843a6d2f4a758fd638b72c594883,
title = "Reconstruction of high-resolution facial image using recursive error back-projection",
abstract = "This paper proposes a new method of reconstructing high-resolution facial image from a low-resolution facial image using a recursive error back-projection of example-based learning. A face is represented by a linear combination of prototypes of shape and texture. With the shape and texture information about the pixels in a given low-resolution facial image, we can estimate optimal coefficients for a linear combination of prototypes of shape and those of texture by solving least square minimization. Then high-resolution facial image can be reconstructed by using the optimal coefficients for linear combination of the high-resolution prototypes. Moreover recursive error back-projection is applied to improve the accuracy of high-resolution reconstruction. An error back-projection is composed of estimation, simulation, and error compensation. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at visual surveillance systems.",
author = "Park, {Jeong Seon} and Lee, {Seong Whan}",
year = "2004",
month = "12",
day = "1",
language = "English",
volume = "3072",
pages = "59--66",
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 - Reconstruction of high-resolution facial image using recursive error back-projection

AU - Park, Jeong Seon

AU - Lee, Seong Whan

PY - 2004/12/1

Y1 - 2004/12/1

N2 - This paper proposes a new method of reconstructing high-resolution facial image from a low-resolution facial image using a recursive error back-projection of example-based learning. A face is represented by a linear combination of prototypes of shape and texture. With the shape and texture information about the pixels in a given low-resolution facial image, we can estimate optimal coefficients for a linear combination of prototypes of shape and those of texture by solving least square minimization. Then high-resolution facial image can be reconstructed by using the optimal coefficients for linear combination of the high-resolution prototypes. Moreover recursive error back-projection is applied to improve the accuracy of high-resolution reconstruction. An error back-projection is composed of estimation, simulation, and error compensation. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at visual surveillance systems.

AB - This paper proposes a new method of reconstructing high-resolution facial image from a low-resolution facial image using a recursive error back-projection of example-based learning. A face is represented by a linear combination of prototypes of shape and texture. With the shape and texture information about the pixels in a given low-resolution facial image, we can estimate optimal coefficients for a linear combination of prototypes of shape and those of texture by solving least square minimization. Then high-resolution facial image can be reconstructed by using the optimal coefficients for linear combination of the high-resolution prototypes. Moreover recursive error back-projection is applied to improve the accuracy of high-resolution reconstruction. An error back-projection is composed of estimation, simulation, and error compensation. The encouraging results of the proposed method show that our method can be used to improve the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at visual surveillance systems.

UR - http://www.scopus.com/inward/record.url?scp=35048842922&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=35048842922&partnerID=8YFLogxK

M3 - Article

VL - 3072

SP - 59

EP - 66

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 -