Topological mappings of video and audio data

Colin Fyfe, Wesam Barbakh, Wei Chuan Ooi, Hanseok Ko

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

We review a new form of self-organizing map which is based on a nonlinear projection of latent points into data space, identical to that performed in the Generative Topographic Mapping (GTM).1 But whereas the GTM is an extension of a mixture of experts, this model is an extension of a product of experts.2 We show visualisation and clustering results on a data set composed of video data of lips uttering 5 Korean vowels. Finally we note that we may dispense with the probabilistic underpinnings of the product of experts and derive the same algorithm as a minimisation of mean squared error between the prototypes and the data. This leads us to suggest a new algorithm which incorporates local and global information in the clustering. Both ot the new algorithms achieve better results than the standard Self-Organizing Map.

Original languageEnglish
Pages (from-to)481-489
Number of pages9
JournalInternational Journal of Neural Systems
Volume18
Issue number6
DOIs
Publication statusPublished - 2008 Dec 1

Fingerprint

Self organizing maps
Visualization

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Topological mappings of video and audio data. / Fyfe, Colin; Barbakh, Wesam; Ooi, Wei Chuan; Ko, Hanseok.

In: International Journal of Neural Systems, Vol. 18, No. 6, 01.12.2008, p. 481-489.

Research output: Contribution to journalArticle

Fyfe, Colin ; Barbakh, Wesam ; Ooi, Wei Chuan ; Ko, Hanseok. / Topological mappings of video and audio data. In: International Journal of Neural Systems. 2008 ; Vol. 18, No. 6. pp. 481-489.
@article{2b54397971284473b5ac5b36a3039ca1,
title = "Topological mappings of video and audio data",
abstract = "We review a new form of self-organizing map which is based on a nonlinear projection of latent points into data space, identical to that performed in the Generative Topographic Mapping (GTM).1 But whereas the GTM is an extension of a mixture of experts, this model is an extension of a product of experts.2 We show visualisation and clustering results on a data set composed of video data of lips uttering 5 Korean vowels. Finally we note that we may dispense with the probabilistic underpinnings of the product of experts and derive the same algorithm as a minimisation of mean squared error between the prototypes and the data. This leads us to suggest a new algorithm which incorporates local and global information in the clustering. Both ot the new algorithms achieve better results than the standard Self-Organizing Map.",
author = "Colin Fyfe and Wesam Barbakh and Ooi, {Wei Chuan} and Hanseok Ko",
year = "2008",
month = "12",
day = "1",
doi = "10.1142/S0129065708001749",
language = "English",
volume = "18",
pages = "481--489",
journal = "International Journal of Neural Systems",
issn = "0129-0657",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "6",

}

TY - JOUR

T1 - Topological mappings of video and audio data

AU - Fyfe, Colin

AU - Barbakh, Wesam

AU - Ooi, Wei Chuan

AU - Ko, Hanseok

PY - 2008/12/1

Y1 - 2008/12/1

N2 - We review a new form of self-organizing map which is based on a nonlinear projection of latent points into data space, identical to that performed in the Generative Topographic Mapping (GTM).1 But whereas the GTM is an extension of a mixture of experts, this model is an extension of a product of experts.2 We show visualisation and clustering results on a data set composed of video data of lips uttering 5 Korean vowels. Finally we note that we may dispense with the probabilistic underpinnings of the product of experts and derive the same algorithm as a minimisation of mean squared error between the prototypes and the data. This leads us to suggest a new algorithm which incorporates local and global information in the clustering. Both ot the new algorithms achieve better results than the standard Self-Organizing Map.

AB - We review a new form of self-organizing map which is based on a nonlinear projection of latent points into data space, identical to that performed in the Generative Topographic Mapping (GTM).1 But whereas the GTM is an extension of a mixture of experts, this model is an extension of a product of experts.2 We show visualisation and clustering results on a data set composed of video data of lips uttering 5 Korean vowels. Finally we note that we may dispense with the probabilistic underpinnings of the product of experts and derive the same algorithm as a minimisation of mean squared error between the prototypes and the data. This leads us to suggest a new algorithm which incorporates local and global information in the clustering. Both ot the new algorithms achieve better results than the standard Self-Organizing Map.

UR - http://www.scopus.com/inward/record.url?scp=58449106238&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58449106238&partnerID=8YFLogxK

U2 - 10.1142/S0129065708001749

DO - 10.1142/S0129065708001749

M3 - Article

C2 - 19145664

AN - SCOPUS:58449106238

VL - 18

SP - 481

EP - 489

JO - International Journal of Neural Systems

JF - International Journal of Neural Systems

SN - 0129-0657

IS - 6

ER -