Presentation | 2013-03-14 Fine-grained classification with various local descriptors and Passive-Aggressive Naoyuki GUNJI, Takayuki HIGUCHI, Koki YASUMOTO, Hiroshi MURAOKA, Yoshitaka USHIKU, Tatsuya HARADA, Yasuo KUNIYOSHI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Recent years, fine-grained classification has been studied intensively. But, various local descriptors and encodings proposed for general object classification have not been applied to fine-grained classification many times. In this work, we evaluate such methods with a dataset, which was used in ImageNet Large Scale Visual Recognition Challenge (ILSVRC)2012 fine-grained classification task. When we use densely sampled SIFT descriptors and Fisher Vector, we achieve 20.5%accuracy. But, when we combine many kinds of local descriptors and utilize bounding boxes, our system yields 40.6%accuracy finally and ranks top in ILSVRC2012 fine-grained classification task. Furthermore, when we compare our system to a state-of-the-art method on a similar dataset, our system outperforms the state-of-the-art method in accuracy. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Large-scale image classification / local descriptor / fine-grained classification / Fisher Vector |
Paper # | PRMU2012-184 |
Date of Issue |
Conference Information | |
Committee | PRMU |
---|---|
Conference Date | 2013/3/7(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) | Fine-grained classification with various local descriptors and Passive-Aggressive |
Sub Title (in English) | |
Keyword(1) | Large-scale image classification |
Keyword(2) | local descriptor |
Keyword(3) | fine-grained classification |
Keyword(4) | Fisher Vector |
1st Author's Name | Naoyuki GUNJI |
1st Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo() |
2nd Author's Name | Takayuki HIGUCHI |
2nd Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
3rd Author's Name | Koki YASUMOTO |
3rd Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
4th Author's Name | Hiroshi MURAOKA |
4th Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
5th Author's Name | Yoshitaka USHIKU |
5th Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
6th Author's Name | Tatsuya HARADA |
6th Author's Affiliation | Graduate School of Information Science and Technology:JST PRESTO, The University of Tokyo |
7th Author's Name | Yasuo KUNIYOSHI |
7th Author's Affiliation | Graduate School of Information Science and Technology, The University of Tokyo |
Date | 2013-03-14 |
Paper # | PRMU2012-184 |
Volume (vol) | vol.112 |
Number (no) | 495 |
Page | pp.pp.- |
#Pages | 6 |
Date of Issue |