Presentation 1993/5/26
Quantization Level increase in Human Face Using Multilayer Neural Network
Yoshinori Kimura, Hiroshi Katayama, Kenji Nakayama,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Quantization level increase in human face images using multilayer neural network(NN)is discussed.Basically speaking,it is impossible to increase quantization level without any other information.In this paper,however,images are limited to human faces.Therefore,a point of this discussion is the following:Is it possible to extract genera feature of human face by using NN? In order to investigate this point,the multilayer NN is trained using 8-level images as inputs,and the same images with 256-levels as targets.Back-propagation algorithm is employed.By using about 100 face images,the NN can transform untrained 8-level face images to 256-level face images.The trained NN holds facility of transforming arbitrary input images to human face like images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Multilayer Neural Network / Face Image / Quantization Level / Image Restoration / Image Transformation
Paper # NC93-2
Date of Issue

Conference Information
Committee NC
Conference Date 1993/5/26(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Quantization Level increase in Human Face Using Multilayer Neural Network
Sub Title (in English)
Keyword(1) Multilayer Neural Network
Keyword(2) Face Image
Keyword(3) Quantization Level
Keyword(4) Image Restoration
Keyword(5) Image Transformation
1st Author's Name Yoshinori Kimura
1st Author's Affiliation Department of Electrical and Computer Engineering,Faculty of Technology,Kanazawa University()
2nd Author's Name Hiroshi Katayama
2nd Author's Affiliation Fujitsu Limited
3rd Author's Name Kenji Nakayama
3rd Author's Affiliation Department of Electrical and Computer Engineering,Faculty of Technology,Kanazawa University
Date 1993/5/26
Paper # NC93-2
Volume (vol) vol.93
Number (no) 67
Page pp.pp.-
#Pages 7
Date of Issue