Parallel-hierarchical neural network for 3D object recognition. Sato, N. & Hagiwara, M. Systems and Computers in Japan, 35(1):1--12, 2004. ![Parallel-hierarchical neural network for 3D object recognition [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex In this paper, the authors propose a parallel-hierarchical neural network that can recognize multiple 3D objects from 2D projection images. The proposed network focuses on a parallel-hierarchical structure and memory-based recognition assistance, which are characteristics of the excellent vision systems that living organisms have, and refers to the neocognitron, which models the parallel-hierarchical structure of vision systems. The amount of calculations is reduced by deleting cells having a low degree of importance based on competition between cells and detecting features that differ for each cell. The network not only can recognize an object, but can also estimate its orientation at the same time. The memory-based recognition assistance is modeled by performing iterative processing during which the input image approaches a learning image based on the orientation estimation result. Weights are determined by sequentially presenting learning images, and no teaching signal is necessary. This is done in a short time since it is not an iterative learning method. The network performance was evaluated by using five objects from COIL-100. Images that were obtained by photographing each object after it was rotated 60° at a time around the vertical axis were used for the learning images. The experiments checked the recognition rates for various 2D projection images that were obtained from the 3D objects. These results verified the effectiveness of the proposed neural network as a 3D object recognition technique. © 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 35(1): 1–12, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10579
@article{sato_parallel-hierarchical_2004,
title = {Parallel-hierarchical neural network for 3D object recognition},
volume = {35},
copyright = {Copyright © 2003 Wiley Periodicals, Inc.},
issn = {1520-684X},
url = {http://onlinelibrary.wiley.com/doi/10.1002/scj.10579/abstract},
doi = {10.1002/scj.10579},
abstract = {In this paper, the authors propose a parallel-hierarchical neural network that can recognize multiple 3D objects from 2D projection images. The proposed network focuses on a parallel-hierarchical structure and memory-based recognition assistance, which are characteristics of the excellent vision systems that living organisms have, and refers to the neocognitron, which models the parallel-hierarchical structure of vision systems. The amount of calculations is reduced by deleting cells having a low degree of importance based on competition between cells and detecting features that differ for each cell. The network not only can recognize an object, but can also estimate its orientation at the same time. The memory-based recognition assistance is modeled by performing iterative processing during which the input image approaches a learning image based on the orientation estimation result. Weights are determined by sequentially presenting learning images, and no teaching signal is necessary. This is done in a short time since it is not an iterative learning method. The network performance was evaluated by using five objects from COIL-100. Images that were obtained by photographing each object after it was rotated 60° at a time around the vertical axis were used for the learning images. The experiments checked the recognition rates for various 2D projection images that were obtained from the 3D objects. These results verified the effectiveness of the proposed neural network as a 3D object recognition technique. © 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 35(1): 1–12, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10579},
language = {en},
number = {1},
urldate = {2015-03-22TZ},
journal = {Systems and Computers in Japan},
author = {Sato, Noriaki and Hagiwara, Masafumi},
year = {2004},
keywords = {3D object recognition, affine transformation recognition, occlusion recognition, vision-based recognition model},
pages = {1--12}
}
Downloads: 0
{"_id":"2jQZzd2oNoaieJ4CJ","bibbaseid":"sato-hagiwara-parallelhierarchicalneuralnetworkfor3dobjectrecognition-2004","downloads":0,"creationDate":"2016-03-16T14:38:26.465Z","title":"Parallel-hierarchical neural network for 3D object recognition","author_short":["Sato, N.","Hagiwara, M."],"year":2004,"bibtype":"article","biburl":"http://bibbase.org/zotero/hogsford","bibdata":{"bibtype":"article","type":"article","title":"Parallel-hierarchical neural network for 3D object recognition","volume":"35","copyright":"Copyright © 2003 Wiley Periodicals, Inc.","issn":"1520-684X","url":"http://onlinelibrary.wiley.com/doi/10.1002/scj.10579/abstract","doi":"10.1002/scj.10579","abstract":"In this paper, the authors propose a parallel-hierarchical neural network that can recognize multiple 3D objects from 2D projection images. The proposed network focuses on a parallel-hierarchical structure and memory-based recognition assistance, which are characteristics of the excellent vision systems that living organisms have, and refers to the neocognitron, which models the parallel-hierarchical structure of vision systems. The amount of calculations is reduced by deleting cells having a low degree of importance based on competition between cells and detecting features that differ for each cell. The network not only can recognize an object, but can also estimate its orientation at the same time. The memory-based recognition assistance is modeled by performing iterative processing during which the input image approaches a learning image based on the orientation estimation result. Weights are determined by sequentially presenting learning images, and no teaching signal is necessary. This is done in a short time since it is not an iterative learning method. The network performance was evaluated by using five objects from COIL-100. Images that were obtained by photographing each object after it was rotated 60° at a time around the vertical axis were used for the learning images. The experiments checked the recognition rates for various 2D projection images that were obtained from the 3D objects. These results verified the effectiveness of the proposed neural network as a 3D object recognition technique. © 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 35(1): 1–12, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10579","language":"en","number":"1","urldate":"2015-03-22TZ","journal":"Systems and Computers in Japan","author":[{"propositions":[],"lastnames":["Sato"],"firstnames":["Noriaki"],"suffixes":[]},{"propositions":[],"lastnames":["Hagiwara"],"firstnames":["Masafumi"],"suffixes":[]}],"year":"2004","keywords":"3D object recognition, affine transformation recognition, occlusion recognition, vision-based recognition model","pages":"1--12","bibtex":"@article{sato_parallel-hierarchical_2004,\n\ttitle = {Parallel-hierarchical neural network for 3D object recognition},\n\tvolume = {35},\n\tcopyright = {Copyright © 2003 Wiley Periodicals, Inc.},\n\tissn = {1520-684X},\n\turl = {http://onlinelibrary.wiley.com/doi/10.1002/scj.10579/abstract},\n\tdoi = {10.1002/scj.10579},\n\tabstract = {In this paper, the authors propose a parallel-hierarchical neural network that can recognize multiple 3D objects from 2D projection images. The proposed network focuses on a parallel-hierarchical structure and memory-based recognition assistance, which are characteristics of the excellent vision systems that living organisms have, and refers to the neocognitron, which models the parallel-hierarchical structure of vision systems. The amount of calculations is reduced by deleting cells having a low degree of importance based on competition between cells and detecting features that differ for each cell. The network not only can recognize an object, but can also estimate its orientation at the same time. The memory-based recognition assistance is modeled by performing iterative processing during which the input image approaches a learning image based on the orientation estimation result. Weights are determined by sequentially presenting learning images, and no teaching signal is necessary. This is done in a short time since it is not an iterative learning method. The network performance was evaluated by using five objects from COIL-100. Images that were obtained by photographing each object after it was rotated 60° at a time around the vertical axis were used for the learning images. The experiments checked the recognition rates for various 2D projection images that were obtained from the 3D objects. These results verified the effectiveness of the proposed neural network as a 3D object recognition technique. © 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 35(1): 1–12, 2004; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.10579},\n\tlanguage = {en},\n\tnumber = {1},\n\turldate = {2015-03-22TZ},\n\tjournal = {Systems and Computers in Japan},\n\tauthor = {Sato, Noriaki and Hagiwara, Masafumi},\n\tyear = {2004},\n\tkeywords = {3D object recognition, affine transformation recognition, occlusion recognition, vision-based recognition model},\n\tpages = {1--12}\n}\n\n","author_short":["Sato, N.","Hagiwara, M."],"key":"sato_parallel-hierarchical_2004","id":"sato_parallel-hierarchical_2004","bibbaseid":"sato-hagiwara-parallelhierarchicalneuralnetworkfor3dobjectrecognition-2004","role":"author","urls":{"Paper":"http://onlinelibrary.wiley.com/doi/10.1002/scj.10579/abstract"},"keyword":["3D object recognition","affine transformation recognition","occlusion recognition","vision-based recognition model"],"downloads":0},"search_terms":["parallel","hierarchical","neural","network","object","recognition","sato","hagiwara"],"keywords":["3d object recognition","affine transformation recognition","occlusion recognition","vision-based recognition model"],"authorIDs":[],"dataSources":["v98BPm7u4WMoReKPh"]}