Presentation 2015-03-04
Fast Bag-of-Feature Generation Featuring Approximate Nearest Image-Patch Search
Kei ITO, Toshihiko YAMASAKI, Kiyoharu AIZAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In Bag-of-Features models, which have been widely used in object recognition, various speed-up approaches have been developed. However the process of extracting local features such as SIFT remains as the bottle neck of processing time. In this paper, we shorten the processing time by skipping the step of extracting local features. We make a huge table of image-patches and IDs from the learning dataset beforehand. When we calculate the ID attached to the query image-patch, we pick it up from the table by approximate nearest neighbor search.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Bag-of-Features / SIFT / Approximate Nearest Neighbor Search
Paper # IMQ2014-53,IE2014-114,MVE2014-101
Date of Issue

Conference Information
Committee MVE
Conference Date 2015/2/24(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 Media Experience and Virtual Environment (MVE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Fast Bag-of-Feature Generation Featuring Approximate Nearest Image-Patch Search
Sub Title (in English)
Keyword(1) Bag-of-Features
Keyword(2) SIFT
Keyword(3) Approximate Nearest Neighbor Search
1st Author's Name Kei ITO
1st Author's Affiliation Department of Information and Communication Engineering, The University of Tokyo()
2nd Author's Name Toshihiko YAMASAKI
2nd Author's Affiliation Department of Information and Communication Engineering, The University of Tokyo
3rd Author's Name Kiyoharu AIZAWA
3rd Author's Affiliation Department of Information and Communication Engineering, The University of Tokyo
Date 2015-03-04
Paper # IMQ2014-53,IE2014-114,MVE2014-101
Volume (vol) vol.114
Number (no) 487
Page pp.pp.-
#Pages 2
Date of Issue