| Presentation | 2005/3/23 Self-Organizing Map of 3D Objects by the Modular Network SOM (mnSOM) Kenji MORISHITA, Kazuhiro TOKUNAGA, Syozo YASUI, Tetsuo FURUKAWA, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A new method proposed in this paper aims to generate a self-organizing map (SOM) of 3D objects by using their 2D projected images. Since it is known that a set of 2D images of a 3D object form various viewpoints forms a 2D manifold, the method needs to be something that can deal with a set of manifolds rather than a set of vectors. For this reason, the present approach resorts to the Modular Network SOM (mnSOM) which was recently proposed by the present authors as a generalization of the conventional SOM. In general, a mnSOM consists of an assembly of functional modules arrayed on a lattice. In this work, Autoassociative Neural Networks (ANNs) and alternatively the basic SOMs are adopted as the modules, so that the mnSOM acequires an ability to represent the group of 2D manifolds. Thus, the architectures and algorithms are introduced herein for the present applications of mnSOM, and some simulation results are presented for a case of artificial 3D objects. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | 3D Object / self-organizing map / mnSOM / neural network module |
| Paper # | NC2004-210 |
| Date of Issue |
| Conference Information | |
| Committee | NC |
|---|---|
| Conference Date | 2005/3/23(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) | Self-Organizing Map of 3D Objects by the Modular Network SOM (mnSOM) |
| Sub Title (in English) | |
| Keyword(1) | 3D Object |
| Keyword(2) | self-organizing map |
| Keyword(3) | mnSOM |
| Keyword(4) | neural network module |
| 1st Author's Name | Kenji MORISHITA |
| 1st Author's Affiliation | Graduate School of Life Science and System Engineering, Kyushu Institute of Technology() |
| 2nd Author's Name | Kazuhiro TOKUNAGA |
| 2nd Author's Affiliation | Graduate School of Life Science and System Engineering, Kyushu Institute of Technology |
| 3rd Author's Name | Syozo YASUI |
| 3rd Author's Affiliation | Graduate School of Life Science and System Engineering, Kyushu Institute of Technology |
| 4th Author's Name | Tetsuo FURUKAWA |
| 4th Author's Affiliation | Graduate School of Life Science and System Engineering, Kyushu Institute of Technology |
| Date | 2005/3/23 |
| Paper # | NC2004-210 |
| Volume (vol) | vol.104 |
| Number (no) | 760 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |