Presentation 1996/2/3
Recognition of Three-Dimensional Resembled Objects Using Conditional Learning Type Neural Network
Katsuhito Motoyoshi, Yoshikazu Miyanaga, Koji Tochinai,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This report proposes a new learning technique using a feed-forward Neural Network to develop the recognition system of three-dimensional objects. The proposed algorithm does not require complicated pre-processing which is usually introduced into many conventional systems. This system can recognize unknown 2-dimensional view images and hence shows effective for the recognition of the resembled 3-dimensional objects. At the same time this report shows a structure of New-Network systems to be suitable for recognizing variety corresponding to various distances and principle movements. Accordingly point of sight realized in this network has two free-parameters, i.e., an angle and the distance.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) condition learning type neural network / add weight coefficient / viewpoint vector and middle unit positioning vector
Paper # NC95-103
Date of Issue

Conference Information
Committee NC
Conference Date 1996/2/3(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) Recognition of Three-Dimensional Resembled Objects Using Conditional Learning Type Neural Network
Sub Title (in English)
Keyword(1) condition learning type neural network
Keyword(2) add weight coefficient
Keyword(3) viewpoint vector and middle unit positioning vector
1st Author's Name Katsuhito Motoyoshi
1st Author's Affiliation Department of Electronic Engineering, Faculty of Engineering, Hokkaido University()
2nd Author's Name Yoshikazu Miyanaga
2nd Author's Affiliation Department of Electronic Engineering, Hokkaido University:IEICE
3rd Author's Name Koji Tochinai
3rd Author's Affiliation Department of Electronic Engineering, Hokkaido University:IEICE
Date 1996/2/3
Paper # NC95-103
Volume (vol) vol.95
Number (no) 506
Page pp.pp.-
#Pages 8
Date of Issue