Generalised nearest feature line for subspace learning

Y. Pang; Y. Yuan; X. Li

doi:10.1049/el:20072176

Generalised nearest feature line for subspace learning

Y. Pang^*, Y. Yuan, X. Li

^*Corresponding author for this work

Computer Science Research Group

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

Abstract

Nearest feature line (NFL) is a powerful tool in pattern recognition, which generalises the representational capacity of available prototypes by using linear interpolation and extrapolation between the feature points. NFL is generalised to subspace learning so that the obtained subspace has desirable discriminating ability. Experiments on face recognition demonstrate its effectiveness.

Original language	English
Pages (from-to)	1079-1080
Number of pages	2
Journal	Electronics letters
Volume	43
Issue number	20
DOIs	https://doi.org/10.1049/el:20072176
Publication status	Published - 8 Oct 2007

Keywords

discriminators
extrapolation
interpolation
software prototyping

Access to Document

10.1049/el:20072176

Cite this

@article{fc24db547e6447d6b3e8faf497aed8d9,

title = "Generalised nearest feature line for subspace learning",

abstract = "Nearest feature line (NFL) is a powerful tool in pattern recognition, which generalises the representational capacity of available prototypes by using linear interpolation and extrapolation between the feature points. NFL is generalised to subspace learning so that the obtained subspace has desirable discriminating ability. Experiments on face recognition demonstrate its effectiveness.",

keywords = "discriminators, extrapolation, interpolation, software prototyping",

author = "Y. Pang and Y. Yuan and X. Li",

year = "2007",

month = oct,

day = "8",

doi = "10.1049/el:20072176",

language = "English",

volume = "43",

pages = "1079--1080",

journal = "Electronics letters",

issn = "0013-5194",

publisher = "IET",

number = "20",

}

TY - JOUR

T1 - Generalised nearest feature line for subspace learning

AU - Pang, Y.

AU - Yuan, Y.

AU - Li, X.

PY - 2007/10/8

Y1 - 2007/10/8

N2 - Nearest feature line (NFL) is a powerful tool in pattern recognition, which generalises the representational capacity of available prototypes by using linear interpolation and extrapolation between the feature points. NFL is generalised to subspace learning so that the obtained subspace has desirable discriminating ability. Experiments on face recognition demonstrate its effectiveness.

AB - Nearest feature line (NFL) is a powerful tool in pattern recognition, which generalises the representational capacity of available prototypes by using linear interpolation and extrapolation between the feature points. NFL is generalised to subspace learning so that the obtained subspace has desirable discriminating ability. Experiments on face recognition demonstrate its effectiveness.

KW - discriminators

KW - extrapolation

KW - interpolation

KW - software prototyping

UR - http://www.scopus.com/inward/record.url?scp=34748847406&partnerID=8YFLogxK

U2 - 10.1049/el:20072176

DO - 10.1049/el:20072176

M3 - Article

AN - SCOPUS:34748847406

SN - 0013-5194

VL - 43

SP - 1079

EP - 1080

JO - Electronics letters

JF - Electronics letters

IS - 20

ER -

Generalised nearest feature line for subspace learning

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this