Sorted locally confined non-negative matrix factorization in face verification

In this paper, we propose a face recognition technique based on modification of Non-Negative Matrix Factorization (NMF) technique, which known as Sorted Locally Confined NMF (SLC-NMF). SLC-NMF used NMF to find non negative basis images, subset of them were selected according to a discriminant factor...

Full description

Bibliographic Details
Main Authors: Teoh, , ABJ, Ngo, , DCL, Neo, , HF
Format: Article
Published: 2005
Subjects:
Online Access:http://shdl.mmu.edu.my/2335/
_version_ 1848790028851871744
author Teoh, , ABJ
Ngo, , DCL
Neo, , HF
author_facet Teoh, , ABJ
Ngo, , DCL
Neo, , HF
author_sort Teoh, , ABJ
building MMU Institutional Repository
collection Online Access
description In this paper, we propose a face recognition technique based on modification of Non-Negative Matrix Factorization (NMF) technique, which known as Sorted Locally Confined NMF (SLC-NMF). SLC-NMF used NMF to find non negative basis images, subset of them were selected according to a discriminant factor and then processed through a series of image processing operation; to yield a set of ideal locally confined salient feature basis images. SLC-NMF illustrates perfectly local salient feature region which effectively realize "recognition by parts" paradigm for face recognition. The best performance is attained by SLC-NMF compare to the PCA, NMF and local NMF, in FERET Face Database.
first_indexed 2025-11-14T18:06:06Z
format Article
id mmu-2335
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:06:06Z
publishDate 2005
recordtype eprints
repository_type Digital Repository
spelling mmu-23352011-08-23T07:36:02Z http://shdl.mmu.edu.my/2335/ Sorted locally confined non-negative matrix factorization in face verification Teoh, , ABJ Ngo, , DCL Neo, , HF TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television In this paper, we propose a face recognition technique based on modification of Non-Negative Matrix Factorization (NMF) technique, which known as Sorted Locally Confined NMF (SLC-NMF). SLC-NMF used NMF to find non negative basis images, subset of them were selected according to a discriminant factor and then processed through a series of image processing operation; to yield a set of ideal locally confined salient feature basis images. SLC-NMF illustrates perfectly local salient feature region which effectively realize "recognition by parts" paradigm for face recognition. The best performance is attained by SLC-NMF compare to the PCA, NMF and local NMF, in FERET Face Database. 2005 Article NonPeerReviewed Teoh, , ABJ and Ngo, , DCL and Neo, , HF (2005) Sorted locally confined non-negative matrix factorization in face verification. 2005 INTERNATIONAL CONFERENCE ON COMMUNICATIONS, CIRCUITS AND SYSTEMS, VOLS 1 AND 2, PROCEEDINGS: VOL 1: COMMUNICATION THEORY AND SYSTEMS - VOL 2: SIGNAL PROCESSING, COMPUTATIONAL INTELLIGENCE, CIRCUITS AND SYSTEMS. pp. 820-824.
spellingShingle TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
Teoh, , ABJ
Ngo, , DCL
Neo, , HF
Sorted locally confined non-negative matrix factorization in face verification
title Sorted locally confined non-negative matrix factorization in face verification
title_full Sorted locally confined non-negative matrix factorization in face verification
title_fullStr Sorted locally confined non-negative matrix factorization in face verification
title_full_unstemmed Sorted locally confined non-negative matrix factorization in face verification
title_short Sorted locally confined non-negative matrix factorization in face verification
title_sort sorted locally confined non-negative matrix factorization in face verification
topic TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
url http://shdl.mmu.edu.my/2335/