Photo-sketch synthesis and recognition based on subspace learning

Bing Xiao; Xinbo Gao; Dacheng Tao; Yuan Yuan; Jie Li

doi:10.1016/j.neucom.2009.10.014

Photo-sketch synthesis and recognition based on subspace learning

Bing Xiao, Xinbo Gao^*, Dacheng Tao, Yuan Yuan, Jie Li

^*Corresponding author for this work

Computer Science Research Group

Research output: Contribution to journal › Article › peer-review

Abstract

This paper aims to reducing difference between sketches and photos by synthesizing sketches from photos, and vice versa, and then performing sketch-sketch/photo-photo recognition with subspace learning based methods. Pseudo-sketch/pseudo-photo patches are synthesized with embedded hidden Markov model. Because these patches are assembled by averaging their overlapping area in most of the local strategy based methods, which leads to blurring effect to the resulted pseudo-sketch/pseudo-photo, we integrate the patches with image quilting. Experiments are carried out to demonstrate that the proposed method is effective to produce pseudo-sketch/pseudo-photo with high quality and achieve promising recognition results.

Original language	English
Pages (from-to)	840-852
Number of pages	13
Journal	Neurocomputing
Volume	73
Issue number	4-6
Early online date	18 Nov 2009
DOIs	https://doi.org/10.1016/j.neucom.2009.10.014
Publication status	Published - Jan 2010

Bibliographical note

Bayesian Networks / Design and Application of Neural Networks and Intelligent Learning Systems (KES 2008 / Bio-inspired Computing: Theories and Applications (BIC-TA 2007)

Keywords

Embedded hidden Markov model
Image quilting
Pseudo-photo
Pseudo-sketch
Sketch-photo recognition

Access to Document

10.1016/j.neucom.2009.10.014

Cite this

@article{6cd91fcc87c646c3b50d63d35a7a060b,

title = "Photo-sketch synthesis and recognition based on subspace learning",

abstract = "This paper aims to reducing difference between sketches and photos by synthesizing sketches from photos, and vice versa, and then performing sketch-sketch/photo-photo recognition with subspace learning based methods. Pseudo-sketch/pseudo-photo patches are synthesized with embedded hidden Markov model. Because these patches are assembled by averaging their overlapping area in most of the local strategy based methods, which leads to blurring effect to the resulted pseudo-sketch/pseudo-photo, we integrate the patches with image quilting. Experiments are carried out to demonstrate that the proposed method is effective to produce pseudo-sketch/pseudo-photo with high quality and achieve promising recognition results.",

keywords = "Embedded hidden Markov model, Image quilting, Pseudo-photo, Pseudo-sketch, Sketch-photo recognition",

author = "Bing Xiao and Xinbo Gao and Dacheng Tao and Yuan Yuan and Jie Li",

note = "Bayesian Networks / Design and Application of Neural Networks and Intelligent Learning Systems (KES 2008 / Bio-inspired Computing: Theories and Applications (BIC-TA 2007)",

year = "2010",

month = jan,

doi = "10.1016/j.neucom.2009.10.014",

language = "English",

volume = "73",

pages = "840--852",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier",

number = "4-6",

}

TY - JOUR

T1 - Photo-sketch synthesis and recognition based on subspace learning

AU - Xiao, Bing

AU - Gao, Xinbo

AU - Tao, Dacheng

AU - Yuan, Yuan

AU - Li, Jie

N1 - Bayesian Networks / Design and Application of Neural Networks and Intelligent Learning Systems (KES 2008 / Bio-inspired Computing: Theories and Applications (BIC-TA 2007)

PY - 2010/1

Y1 - 2010/1

N2 - This paper aims to reducing difference between sketches and photos by synthesizing sketches from photos, and vice versa, and then performing sketch-sketch/photo-photo recognition with subspace learning based methods. Pseudo-sketch/pseudo-photo patches are synthesized with embedded hidden Markov model. Because these patches are assembled by averaging their overlapping area in most of the local strategy based methods, which leads to blurring effect to the resulted pseudo-sketch/pseudo-photo, we integrate the patches with image quilting. Experiments are carried out to demonstrate that the proposed method is effective to produce pseudo-sketch/pseudo-photo with high quality and achieve promising recognition results.

AB - This paper aims to reducing difference between sketches and photos by synthesizing sketches from photos, and vice versa, and then performing sketch-sketch/photo-photo recognition with subspace learning based methods. Pseudo-sketch/pseudo-photo patches are synthesized with embedded hidden Markov model. Because these patches are assembled by averaging their overlapping area in most of the local strategy based methods, which leads to blurring effect to the resulted pseudo-sketch/pseudo-photo, we integrate the patches with image quilting. Experiments are carried out to demonstrate that the proposed method is effective to produce pseudo-sketch/pseudo-photo with high quality and achieve promising recognition results.

KW - Embedded hidden Markov model

KW - Image quilting

KW - Pseudo-photo

KW - Pseudo-sketch

KW - Sketch-photo recognition

UR - http://www.scopus.com/inward/record.url?scp=75749156904&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2009.10.014

DO - 10.1016/j.neucom.2009.10.014

M3 - Article

AN - SCOPUS:75749156904

SN - 0925-2312

VL - 73

SP - 840

EP - 852

JO - Neurocomputing

JF - Neurocomputing

IS - 4-6

ER -

Photo-sketch synthesis and recognition based on subspace learning

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Incremental tensor biased discriminant analysis: a new color-based visual tracking method

No-reference image quality assessment in contourlet domain

Outlier-resisting graph embedding

Cite this