Presentation 2009-10-22
Content-Based Image Retrieval using SIFT based on Illumination Invariant Features
Masaki KOBAYASHI, Keisuke KAMEYAMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recently, there has been an increasing need for Content-Based Image Retrieval (CBIR), especially the partial image retrieval system, which retrieves partial correspondences in the image database. However, the apparent color of objects are strongly influenced by the illumination variation, and this may affect retrieval results adversely. In this work, we proposed a framework which can find partial matchings by using illumination invariant local features (CSIFT), and achieved the partial image retrieval robust to apparent color changes. Additionally, we experimented in searching partial similarities for queries, such as logos and trademarks, and verified robustness to apparent color changes.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) CBIR / Partial Image Retrieval / Illumination Invariance / Local Feature / SIFT / RANSAC / Homography
Paper # PRMU2009-71
Date of Issue

Conference Information
Committee PRMU
Conference Date 2009/10/15(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 Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Content-Based Image Retrieval using SIFT based on Illumination Invariant Features
Sub Title (in English)
Keyword(1) CBIR
Keyword(2) Partial Image Retrieval
Keyword(3) Illumination Invariance
Keyword(4) Local Feature
Keyword(5) SIFT
Keyword(6) RANSAC
Keyword(7) Homography
1st Author's Name Masaki KOBAYASHI
1st Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba()
2nd Author's Name Keisuke KAMEYAMA
2nd Author's Affiliation Graduate School of Systems and Information Engineering, University of Tsukuba
Date 2009-10-22
Paper # PRMU2009-71
Volume (vol) vol.109
Number (no) 249
Page pp.pp.-
#Pages 6
Date of Issue