Resolution enhancement of facial image based on topdown learning

Jeong Seon Park; Seong Whan Lee

doi:10.1145/982452.982460

Resolution enhancement of facial image based on topdown learning

Jeong Seon Park, Seong Whan Lee

Department of Brain and Cognitive Engineering

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

8 Citations (Scopus)

Abstract

This paper proposes a new method of synthesizing a highresolution facial image from a low-resolution facial image based on top-down 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 synthesized by using the optimal coefficients for linear combination of the high-resolution prototypes. The encouraging results of the proposed method show that our method can be used to increase the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at surveillance systems.

Original language	English
Title of host publication	1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003
Publisher	Association for Computing Machinery, Inc
Pages	59-64
Number of pages	6
ISBN (Electronic)	158113780X, 9781581137804
DOIs	https://doi.org/10.1145/982452.982460
Publication status	Published - 2003 Nov 2
Event	1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003 - Berkeley, United States Duration: 2003 Nov 2 → 2003 Nov 8

Other

Other	1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003
Country	United States
City	Berkeley
Period	03/11/2 → 03/11/8

Keywords

Face recognition
Face reconstruction
Low-resolution
Morphable face model

ASJC Scopus subject areas

Signal Processing
Computer Networks and Communications

Access to Document

10.1145/982452.982460

Fingerprint Dive into the research topics of 'Resolution enhancement of facial image based on topdown learning'. Together they form a unique fingerprint.

Cite this

@inproceedings{54326ed87d85420994361b6e875ba2df,

title = "Resolution enhancement of facial image based on topdown learning",

abstract = "This paper proposes a new method of synthesizing a highresolution facial image from a low-resolution facial image based on top-down 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 synthesized by using the optimal coefficients for linear combination of the high-resolution prototypes. The encouraging results of the proposed method show that our method can be used to increase the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at surveillance systems.",

keywords = "Face recognition, Face reconstruction, Low-resolution, Morphable face model",

author = "Park, {Jeong Seon} and Lee, {Seong Whan}",

year = "2003",

month = nov,

day = "2",

doi = "10.1145/982452.982460",

language = "English",

pages = "59--64",

booktitle = "1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003",

publisher = "Association for Computing Machinery, Inc",

note = "1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003 ; Conference date: 02-11-2003 Through 08-11-2003",

}

TY - GEN

T1 - Resolution enhancement of facial image based on topdown learning

AU - Park, Jeong Seon

AU - Lee, Seong Whan

PY - 2003/11/2

Y1 - 2003/11/2

N2 - This paper proposes a new method of synthesizing a highresolution facial image from a low-resolution facial image based on top-down 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 synthesized by using the optimal coefficients for linear combination of the high-resolution prototypes. The encouraging results of the proposed method show that our method can be used to increase the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at surveillance systems.

AB - This paper proposes a new method of synthesizing a highresolution facial image from a low-resolution facial image based on top-down 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 synthesized by using the optimal coefficients for linear combination of the high-resolution prototypes. The encouraging results of the proposed method show that our method can be used to increase the performance of the face recognition by applying our method to enhance the low-resolution facial images captured at surveillance systems.

KW - Face recognition

KW - Face reconstruction

KW - Low-resolution

KW - Morphable face model

UR - http://www.scopus.com/inward/record.url?scp=84968768275&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84968768275&partnerID=8YFLogxK

U2 - 10.1145/982452.982460

DO - 10.1145/982452.982460

M3 - Conference contribution

AN - SCOPUS:84968768275

SP - 59

EP - 64

BT - 1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003

PB - Association for Computing Machinery, Inc

T2 - 1st ACM SIGMM International Workshop on Video Surveillance, IWVS 2003

Y2 - 2 November 2003 through 8 November 2003

ER -