Presentation 2015-03-19
Log-normal Distribution of Objects' Size in Images and Its Applications to Object Detection : Comparing General Images and Food Images
Shota HORIGUCHI, Kiyoharu AIZAWA, Makoto OGAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) When detecting objects in images by which classifying many location hypotheses, it is necessary to define aspect ratio and scale of detection window in advance. In this paper, we construct a model of size distribution of objects in images, and revealed that the size of foods in images taken for recording dietary follows a log-normal distribution. We apply this characteristic to defining parameters of selective search, the method of generating object hypotheses, and result in high Mean Average Best Overlap in spite of small number of object hypotheses when using FoodLog image dataset.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) food image / object detection / log-normal distribution / selective search
Paper # BioX2014-61,PRMU2014-181
Date of Issue

Conference Information
Committee PRMU
Conference Date 2015/3/12(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) Log-normal Distribution of Objects' Size in Images and Its Applications to Object Detection : Comparing General Images and Food Images
Sub Title (in English)
Keyword(1) food image
Keyword(2) object detection
Keyword(3) log-normal distribution
Keyword(4) selective search
1st Author's Name Shota HORIGUCHI
1st Author's Affiliation Dept. of Electronics and Information Science, The University of Tokyo()
2nd Author's Name Kiyoharu AIZAWA
2nd Author's Affiliation Dept. of Information and Communication Eng., The University of Tokyo
3rd Author's Name Makoto OGAWA
3rd Author's Affiliation foo.log, Inc.
Date 2015-03-19
Paper # BioX2014-61,PRMU2014-181
Volume (vol) vol.114
Number (no) 521
Page pp.pp.-
#Pages 6
Date of Issue