Presentation 2006-09-08
Experimental Investigation of Relation Between Near Neighbor Search Methods for Feature Vectors and Efficiency of Object Recognition
Kazuto NOGUCHI, Tomohiro NAKAI, Koichi KISE, Masakazu IWAMURA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Efficiency of object recognition methods using local descriptors such as SIFT and PCA-SIFT depends largely on the speed of matching between feature vectors since images are described by a large number of feature vectors. Because the matching is considered to be "nearest neighbor (NN) search" of feature vectors, the problem is paraphrased by "how to make the NN search efficient". For the object recognition, it is required that the number of incorrect matching does not exceed that of correct matching. In other words, a certain number of incorrect matching is acceptable. This observation allows us to make NN search more efficient using approximate NN search with reduced distance calculation. For this purpose, we propose two methods: one is to eliminate feature vectors that require a number of distance calculations. The other is to use no distance calculation. From experimental results with 10,000 database images and 2,000 query images, it is shown that the proposed method is two to three times efficient as compared to a method using ANN and can achieve, recognition rate of 98% with 8.3 ms/query.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Object recognition / Approxiate nearest neighbor search / PCA-SIFT / ANN / LSH / Hash
Paper # PRMU2006-68
Date of Issue

Conference Information
Committee PRMU
Conference Date 2006/9/1(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) Experimental Investigation of Relation Between Near Neighbor Search Methods for Feature Vectors and Efficiency of Object Recognition
Sub Title (in English)
Keyword(1) Object recognition
Keyword(2) Approxiate nearest neighbor search
Keyword(3) PCA-SIFT
Keyword(4) ANN
Keyword(5) LSH
Keyword(6) Hash
1st Author's Name Kazuto NOGUCHI
1st Author's Affiliation College of Engineering, Osaka Prefecture University()
2nd Author's Name Tomohiro NAKAI
2nd Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
3rd Author's Name Koichi KISE
3rd Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
4th Author's Name Masakazu IWAMURA
4th Author's Affiliation Graduate School of Engineering, Osaka Prefecture University
Date 2006-09-08
Paper # PRMU2006-68
Volume (vol) vol.106
Number (no) 229
Page pp.pp.-
#Pages 8
Date of Issue