Neighbourhood Discriminant Locally Linear Embedding in Face Recognition

Face images are often very high-dimensional and complex. However, the actual underlying structure can be characterized by a small number of features. Hence, locally linear embedding (LLE) is proposed as a nonlinear dimension reduction technique to deal this problem. LLE learns the intrinsic manifold...

Full description

Bibliographic Details
Main Authors: Han, Pang Ying, Jin, Andrew Teoh Beng, Kiong, Wong Eng
Format: Conference or Workshop Item
Published: 2008
Subjects:
Online Access:http://shdl.mmu.edu.my/2824/
_version_ 1848790159460401152
author Han, Pang Ying
Jin, Andrew Teoh Beng
Kiong, Wong Eng
author_facet Han, Pang Ying
Jin, Andrew Teoh Beng
Kiong, Wong Eng
author_sort Han, Pang Ying
building MMU Institutional Repository
collection Online Access
description Face images are often very high-dimensional and complex. However, the actual underlying structure can be characterized by a small number of features. Hence, locally linear embedding (LLE) is proposed as a nonlinear dimension reduction technique to deal this problem. LLE learns the intrinsic manifold embedded in the high dimensional ambient space by minimizing the global reconstruction error of the neighbourhood in the data set. LLE is popular in analyzing face images with different poses, illuminations or facial expressions for one subject class. It is developed based on the assumption that data that is distributed on a single manifold is having the same class label; hence the process of neighborhood selection is non class-specific. However, this is inappropriate to face recognition as face recognition learns in multiple manifolds where each representing data on one specific class. Here, we modify, the original LLE by embedding prior class information in the process of neighborhood selection. Experimental results demonstrate that our technique consistently outperforms the original LLE in ORL, PIE and FRGC databases.
first_indexed 2025-11-14T18:08:11Z
format Conference or Workshop Item
id mmu-2824
institution Multimedia University
institution_category Local University
last_indexed 2025-11-14T18:08:11Z
publishDate 2008
recordtype eprints
repository_type Digital Repository
spelling mmu-28242011-09-19T03:25:11Z http://shdl.mmu.edu.my/2824/ Neighbourhood Discriminant Locally Linear Embedding in Face Recognition Han, Pang Ying Jin, Andrew Teoh Beng Kiong, Wong Eng T Technology (General) QA75.5-76.95 Electronic computers. Computer science Face images are often very high-dimensional and complex. However, the actual underlying structure can be characterized by a small number of features. Hence, locally linear embedding (LLE) is proposed as a nonlinear dimension reduction technique to deal this problem. LLE learns the intrinsic manifold embedded in the high dimensional ambient space by minimizing the global reconstruction error of the neighbourhood in the data set. LLE is popular in analyzing face images with different poses, illuminations or facial expressions for one subject class. It is developed based on the assumption that data that is distributed on a single manifold is having the same class label; hence the process of neighborhood selection is non class-specific. However, this is inappropriate to face recognition as face recognition learns in multiple manifolds where each representing data on one specific class. Here, we modify, the original LLE by embedding prior class information in the process of neighborhood selection. Experimental results demonstrate that our technique consistently outperforms the original LLE in ORL, PIE and FRGC databases. 2008-08 Conference or Workshop Item NonPeerReviewed Han, Pang Ying and Jin, Andrew Teoh Beng and Kiong, Wong Eng (2008) Neighbourhood Discriminant Locally Linear Embedding in Face Recognition. In: 5th International Conference on Computer Graphics, Imaging and Visualization (CGIV), 26-28 August 2008, Penang, MALAYSIA. http://dx.doi.org/10.1109/CGIV.2008.63 doi:10.1109/CGIV.2008.63 doi:10.1109/CGIV.2008.63
spellingShingle T Technology (General)
QA75.5-76.95 Electronic computers. Computer science
Han, Pang Ying
Jin, Andrew Teoh Beng
Kiong, Wong Eng
Neighbourhood Discriminant Locally Linear Embedding in Face Recognition
title Neighbourhood Discriminant Locally Linear Embedding in Face Recognition
title_full Neighbourhood Discriminant Locally Linear Embedding in Face Recognition
title_fullStr Neighbourhood Discriminant Locally Linear Embedding in Face Recognition
title_full_unstemmed Neighbourhood Discriminant Locally Linear Embedding in Face Recognition
title_short Neighbourhood Discriminant Locally Linear Embedding in Face Recognition
title_sort neighbourhood discriminant locally linear embedding in face recognition
topic T Technology (General)
QA75.5-76.95 Electronic computers. Computer science
url http://shdl.mmu.edu.my/2824/
http://shdl.mmu.edu.my/2824/
http://shdl.mmu.edu.my/2824/