Presentation 2005/2/18
Matching Words and Image Segments for Semantic Retrieval
Andrea KUTICS,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a new method for retrieving images. A user's similarity interpretation is subjective and a similarity model's interpretation is objective. This method combines textual and object-based visual features to decrease this difference. It uses a novel multi-scale segmentation framework to detect prominent objects in an image. These objects are grouped depending on their visual features and mapped to related words obtained from psychophysical studies. Then, a hierarchy of words expressing higher-level meaning is determined. This is based on natural language processing and user evaluation. Experiments were carried out on a large set of natural images. These showed higher retrieval precision in terms of estimating user retrieval semantics obtained via this two-layer word association. Various query specifications and options were also supported
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Retrieval / Semantics / Multi-modal / Segmentation
Paper # NLC2004-127,PRMU2004-209
Date of Issue

Conference Information
Committee NLC
Conference Date 2005/2/18(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 Natural Language Understanding and Models of Communication (NLC)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Matching Words and Image Segments for Semantic Retrieval
Sub Title (in English)
Keyword(1) Image Retrieval
Keyword(2) Semantics
Keyword(3) Multi-modal
Keyword(4) Segmentation
1st Author's Name Andrea KUTICS
1st Author's Affiliation School of Media Science, Tokyo University of Technology()
Date 2005/2/18
Paper # NLC2004-127,PRMU2004-209
Volume (vol) vol.104
Number (no) 668
Page pp.pp.-
#Pages 6
Date of Issue