Presentation 2001/2/2
3D shape representation by a neural network and reconstruction of a 3D shape from its multiple views
Masayoshi Ohno, Itsuo Kumazawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we present a 3D shape modeling method using the learning capability of multi-layer neural network. The network includes polygon parameters and view parameters as its connection weights. It generates projected images for specied view points. The learning capability is used to update the polygon parameters and view parameters so that the camera parameters are automatically estimated and the computed projection images would approximate observed images for corresponding view points. Color information is effectively introduced so that the correct 3D shape is reconstructed from a small number of views.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) 3D reconstruction / Multiple views / Camera calibration / Neural network / Error back propagation
Paper # NC2000-98
Date of Issue

Conference Information
Committee NC
Conference Date 2001/2/2(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) 3D shape representation by a neural network and reconstruction of a 3D shape from its multiple views
Sub Title (in English)
Keyword(1) 3D reconstruction
Keyword(2) Multiple views
Keyword(3) Camera calibration
Keyword(4) Neural network
Keyword(5) Error back propagation
1st Author's Name Masayoshi Ohno
1st Author's Affiliation Department of Computer Science Graduate School of Information Science and Engineering Tokyo Institute of Technology()
2nd Author's Name Itsuo Kumazawa
2nd Author's Affiliation Department of Computer Science Graduate School of Information Science and Engineering Tokyo Institute of Technology
Date 2001/2/2
Paper # NC2000-98
Volume (vol) vol.100
Number (no) 618
Page pp.pp.-
#Pages 8
Date of Issue