Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency

Pooyan Fazli; Ara Bedrosian

Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency

Pooyan Fazli
, Ara Bedrosian

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The objective of this paper is to study the existing methods for unsupervised object recognition and image categorization and propose a model that can learn directly from the output of image search engines, e.g. Google Images, bypassing the need to manually collect large quantities of training data. This model can then be used to refine the quality of the image search, or to search through other sources of images. This integrated scheme has been implemented and optimized to be used in The Semantic Robot Vision Challenge as a new test-bed for research in the areas of image understanding and knowledge retrieval in large unstructured image databases.

Original language	English (US)
Title of host publication	Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008
Publisher	AAAI press
Pages	1796-1797
Number of pages	2
ISBN (Electronic)	9781577353683
State	Published - 2008
Externally published	Yes
Event	23rd AAAI Conference on Artificial Intelligence, AAAI 2008 - Chicago, United States Duration: Jul 13 2008 → Jul 17 2008

Publication series

Name	Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008

Conference

Conference	23rd AAAI Conference on Artificial Intelligence, AAAI 2008
Country/Territory	United States
City	Chicago
Period	7/13/08 → 7/17/08

ASJC Scopus subject areas

Artificial Intelligence

Cite this

Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency. / Fazli, Pooyan; Bedrosian, Ara.
Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008. AAAI press, 2008. p. 1796-1797 (Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Fazli, P & Bedrosian, A 2008, Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency. in Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008. Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008, AAAI press, pp. 1796-1797, 23rd AAAI Conference on Artificial Intelligence, AAAI 2008, Chicago, United States, 7/13/08.

@inproceedings{4756791864fc4851bcb7780e79a164f6,

title = "Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency",

abstract = "The objective of this paper is to study the existing methods for unsupervised object recognition and image categorization and propose a model that can learn directly from the output of image search engines, e.g. Google Images, bypassing the need to manually collect large quantities of training data. This model can then be used to refine the quality of the image search, or to search through other sources of images. This integrated scheme has been implemented and optimized to be used in The Semantic Robot Vision Challenge as a new test-bed for research in the areas of image understanding and knowledge retrieval in large unstructured image databases.",

author = "Pooyan Fazli and Ara Bedrosian",

note = "Publisher Copyright: Copyright {\textcopyright} 2008, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.; 23rd AAAI Conference on Artificial Intelligence, AAAI 2008 ; Conference date: 13-07-2008 Through 17-07-2008",

year = "2008",

language = "English (US)",

series = "Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008",

publisher = "AAAI press",

pages = "1796--1797",

booktitle = "Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008",

}

TY - GEN

T1 - Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency

AU - Fazli, Pooyan

AU - Bedrosian, Ara

PY - 2008

Y1 - 2008

N2 - The objective of this paper is to study the existing methods for unsupervised object recognition and image categorization and propose a model that can learn directly from the output of image search engines, e.g. Google Images, bypassing the need to manually collect large quantities of training data. This model can then be used to refine the quality of the image search, or to search through other sources of images. This integrated scheme has been implemented and optimized to be used in The Semantic Robot Vision Challenge as a new test-bed for research in the areas of image understanding and knowledge retrieval in large unstructured image databases.

AB - The objective of this paper is to study the existing methods for unsupervised object recognition and image categorization and propose a model that can learn directly from the output of image search engines, e.g. Google Images, bypassing the need to manually collect large quantities of training data. This model can then be used to refine the quality of the image search, or to search through other sources of images. This integrated scheme has been implemented and optimized to be used in The Semantic Robot Vision Challenge as a new test-bed for research in the areas of image understanding and knowledge retrieval in large unstructured image databases.

UR - https://www.scopus.com/pages/publications/85167397028

UR - https://www.scopus.com/pages/publications/85167397028#tab=citedBy

M3 - Conference contribution

T3 - Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008

SP - 1796

EP - 1797

BT - Proceedings of the 23rd AAAI Conference on Artificial Intelligence, AAAI 2008

PB - AAAI press

T2 - 23rd AAAI Conference on Artificial Intelligence, AAAI 2008

Y2 - 13 July 2008 through 17 July 2008

ER -

Unsupervised Categorization (Filtering) of Google Images Based on Visual Consistency

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this