Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map

K. W. Chan; K. L. Chan

doi:10.1109/ICNN.1995.488959

Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map

^*Corresponding author for this work

Department of Electrical Engineering

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

Abstract

Reference Neighbor Search (RNS) is a new technique for fast searching of vector quantization (VQ). However, the optimal performance is not guaranteed and the performance is greatly affected by the selection of reference point. In this research, we employed the Kohonen Self Organized Feature Map to generate codebook of high degree of neighborhood and Multi-Layer Perceptron (MLP) neural network to adaptively predict the reference. The predicted reference is closer to the input, thus the search distance will be reduced. Together with multiple queues and a look up table, the number of searches is significantly reduced while maintaining optimal performance. © 1995 IEEE

Original language	English
Title of host publication	Proceedings of ICNN'95 - International Conference on Neural Networks
Publisher	IEEE
Pages	1898-1902
Volume	4
ISBN (Print)	0-7803-2768-3
DOIs	https://doi.org/10.1109/ICNN.1995.488959
Publication status	Published - Nov 1995
Event	1995 IEEE International Conference on Neural Networks (ICNN'95) - Perth, Australia Duration: 27 Nov 1995 → 1 Dec 1995

Conference

Conference	1995 IEEE International Conference on Neural Networks (ICNN'95)
Place	Australia
City	Perth
Period	27/11/95 → 1/12/95

Access Output

10.1109/ICNN.1995.488959

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{edca9e400c68443abd2bff8106885611,

title = "Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map",

abstract = "Reference Neighbor Search (RNS) is a new technique for fast searching of vector quantization (VQ). However, the optimal performance is not guaranteed and the performance is greatly affected by the selection of reference point. In this research, we employed the Kohonen Self Organized Feature Map to generate codebook of high degree of neighborhood and Multi-Layer Perceptron (MLP) neural network to adaptively predict the reference. The predicted reference is closer to the input, thus the search distance will be reduced. Together with multiple queues and a look up table, the number of searches is significantly reduced while maintaining optimal performance. {\textcopyright} 1995 IEEE",

author = "Chan, {K. W.} and Chan, {K. L.}",

year = "1995",

month = nov,

doi = "10.1109/ICNN.1995.488959",

language = "English",

isbn = "0-7803-2768-3",

volume = "4",

pages = "1898--1902",

booktitle = "Proceedings of ICNN'95 - International Conference on Neural Networks",

publisher = "IEEE",

address = "United States",

note = "1995 IEEE International Conference on Neural Networks (ICNN'95) ; Conference date: 27-11-1995 Through 01-12-1995",

}

Chan, KW & Chan, KL 1995, Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map. in Proceedings of ICNN'95 - International Conference on Neural Networks. vol. 4, IEEE, pp. 1898-1902, 1995 IEEE International Conference on Neural Networks (ICNN'95), Perth, Australia, 27/11/95. https://doi.org/10.1109/ICNN.1995.488959

Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map. / Chan, K. W.; Chan, K. L.
Proceedings of ICNN'95 - International Conference on Neural Networks. Vol. 4 IEEE, 1995. p. 1898-1902.

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map

AU - Chan, K. W.

AU - Chan, K. L.

PY - 1995/11

Y1 - 1995/11

N2 - Reference Neighbor Search (RNS) is a new technique for fast searching of vector quantization (VQ). However, the optimal performance is not guaranteed and the performance is greatly affected by the selection of reference point. In this research, we employed the Kohonen Self Organized Feature Map to generate codebook of high degree of neighborhood and Multi-Layer Perceptron (MLP) neural network to adaptively predict the reference. The predicted reference is closer to the input, thus the search distance will be reduced. Together with multiple queues and a look up table, the number of searches is significantly reduced while maintaining optimal performance. © 1995 IEEE

AB - Reference Neighbor Search (RNS) is a new technique for fast searching of vector quantization (VQ). However, the optimal performance is not guaranteed and the performance is greatly affected by the selection of reference point. In this research, we employed the Kohonen Self Organized Feature Map to generate codebook of high degree of neighborhood and Multi-Layer Perceptron (MLP) neural network to adaptively predict the reference. The predicted reference is closer to the input, thus the search distance will be reduced. Together with multiple queues and a look up table, the number of searches is significantly reduced while maintaining optimal performance. © 1995 IEEE

UR - http://www.scopus.com/inward/record.url?scp=0029521623&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-0029521623&origin=recordpage

U2 - 10.1109/ICNN.1995.488959

DO - 10.1109/ICNN.1995.488959

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 0-7803-2768-3

VL - 4

SP - 1898

EP - 1902

BT - Proceedings of ICNN'95 - International Conference on Neural Networks

PB - IEEE

T2 - 1995 IEEE International Conference on Neural Networks (ICNN'95)

Y2 - 27 November 1995 through 1 December 1995

ER -

Multi-reference neighborhood search for vector quantization by neural network prediction and self-organized feature map

Abstract

Conference

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this