Presentation | 2007-06-28 Dynamic Visual Pattern Mining Based on Web Information Yongqing SUN, Satoshi SHIMADA, MASASHI Morimoto, Yukinobu Taniguchi, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper, we propose a novel solution to automatically build visual concept models exploring web image resources. The only input required is one simple query-word to initialize the modeling system. Regarding the query word, the proposed system first extracts its popular semantic concepts using web text information. Then web images associated with each semantic concept are crawled from the web image resources. At last common visual pattern is mined effectively from the web images with the proposed learning algorithm, in which positive and negative samples of good quality are selected from the noisy web images since the quality of learning algorithms is strongly related to the selection of training samples, The proposed solution enables unlimited, dynamic visual concept modeling, and is robust and reliable for generic images. Experimental results demonstrate the effectiveness of the proposed approach. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Web image mining / Image clustering and retrieval / Unsupervised learning / Dynamic visual pattern |
Paper # | DE2007-3,PRMU2007-29 |
Date of Issue |
Conference Information | |
Committee | DE |
---|---|
Conference Date | 2007/6/21(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 | Data Engineering (DE) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Dynamic Visual Pattern Mining Based on Web Information |
Sub Title (in English) | |
Keyword(1) | Web image mining |
Keyword(2) | Image clustering and retrieval |
Keyword(3) | Unsupervised learning |
Keyword(4) | Dynamic visual pattern |
1st Author's Name | Yongqing SUN |
1st Author's Affiliation | NTT Cyber Solutions Laboratories() |
2nd Author's Name | Satoshi SHIMADA |
2nd Author's Affiliation | NTT Cyber Solutions Laboratories |
3rd Author's Name | MASASHI Morimoto |
3rd Author's Affiliation | NTT Resonant Inc. |
4th Author's Name | Yukinobu Taniguchi |
4th Author's Affiliation | NTT Cyber Solutions Laboratories |
Date | 2007-06-28 |
Paper # | DE2007-3,PRMU2007-29 |
Volume (vol) | vol.107 |
Number (no) | 114 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |