Image-to-image retrieval using computationally learned bases and color information

Yasuo Matsuyama; Fuminori Ohashi; Fumiaki Horiike; Tomohiro Nakamura; Shun'ichi Honma; Naoto Katsumata; Yuuki Hoshino

doi:10.1109/IJCNN.2007.4371015

Image-to-image retrieval using computationally learned bases and color information

Yasuo Matsuyama^*, Fuminori Ohashi, Fumiaki Horiike, Tomohiro Nakamura, Shun'ichi Honma, Naoto Katsumata, Yuuki Hoshino

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

New methods for joint compression and Image-to-Image retrieval (I2I retrieval) are presented. The novelty exists in the usage of computationally learned image bases besides color distributions. The bases are obtained by the Principal Component Analysis and/or the Independent Component Analysis. On the image compression, PCA and ICA outperform the JPEG's DCT. This superiority holds even if the bases and superposition coefficients are quantized and encoded. On the I2I retrieval, the precision-recall curve is used to measure the performance. It is found that adding the basis information always increases the baseline ability of the color information. Besides the retrieval evaluation, a unified image format called RIM (Retrieval-aware IMage format) for effective packing of codewords including bases is specified. Furthermore, an image search viewer called Wisvi (Waseda Image Search VIewer) is developed and exploited. A β-version of all source codes can be down-loaded from a web site given in the text.

Original language	English
Title of host publication	IEEE International Conference on Neural Networks - Conference Proceedings
Pages	546-551
Number of pages	6
DOIs	https://doi.org/10.1109/IJCNN.2007.4371015
Publication status	Published - 2007
Event	2007 International Joint Conference on Neural Networks, IJCNN 2007 - Orlando, FL Duration: 2007 Aug 12 → 2007 Aug 17

Other

Other	2007 International Joint Conference on Neural Networks, IJCNN 2007
City	Orlando, FL
Period	07/8/12 → 07/8/17

ASJC Scopus subject areas

Software

Access to Document

10.1109/IJCNN.2007.4371015

Cite this

Matsuyama, Y, Ohashi, F, Horiike, F, Nakamura, T, Honma, S, Katsumata, N & Hoshino, Y 2007, Image-to-image retrieval using computationally learned bases and color information. in IEEE International Conference on Neural Networks - Conference Proceedings., 4371015, pp. 546-551, 2007 International Joint Conference on Neural Networks, IJCNN 2007, Orlando, FL, 07/8/12. https://doi.org/10.1109/IJCNN.2007.4371015

@inproceedings{c19fe577783d45369c7ac51cc9d54035,

title = "Image-to-image retrieval using computationally learned bases and color information",

abstract = "New methods for joint compression and Image-to-Image retrieval (I2I retrieval) are presented. The novelty exists in the usage of computationally learned image bases besides color distributions. The bases are obtained by the Principal Component Analysis and/or the Independent Component Analysis. On the image compression, PCA and ICA outperform the JPEG's DCT. This superiority holds even if the bases and superposition coefficients are quantized and encoded. On the I2I retrieval, the precision-recall curve is used to measure the performance. It is found that adding the basis information always increases the baseline ability of the color information. Besides the retrieval evaluation, a unified image format called RIM (Retrieval-aware IMage format) for effective packing of codewords including bases is specified. Furthermore, an image search viewer called Wisvi (Waseda Image Search VIewer) is developed and exploited. A β-version of all source codes can be down-loaded from a web site given in the text.",

author = "Yasuo Matsuyama and Fuminori Ohashi and Fumiaki Horiike and Tomohiro Nakamura and Shun'ichi Honma and Naoto Katsumata and Yuuki Hoshino",

year = "2007",

doi = "10.1109/IJCNN.2007.4371015",

language = "English",

isbn = "142441380X",

pages = "546--551",

booktitle = "IEEE International Conference on Neural Networks - Conference Proceedings",

note = "2007 International Joint Conference on Neural Networks, IJCNN 2007 ; Conference date: 12-08-2007 Through 17-08-2007",

}

TY - GEN

T1 - Image-to-image retrieval using computationally learned bases and color information

AU - Matsuyama, Yasuo

AU - Ohashi, Fuminori

AU - Horiike, Fumiaki

AU - Nakamura, Tomohiro

AU - Honma, Shun'ichi

AU - Katsumata, Naoto

AU - Hoshino, Yuuki

PY - 2007

Y1 - 2007

N2 - New methods for joint compression and Image-to-Image retrieval (I2I retrieval) are presented. The novelty exists in the usage of computationally learned image bases besides color distributions. The bases are obtained by the Principal Component Analysis and/or the Independent Component Analysis. On the image compression, PCA and ICA outperform the JPEG's DCT. This superiority holds even if the bases and superposition coefficients are quantized and encoded. On the I2I retrieval, the precision-recall curve is used to measure the performance. It is found that adding the basis information always increases the baseline ability of the color information. Besides the retrieval evaluation, a unified image format called RIM (Retrieval-aware IMage format) for effective packing of codewords including bases is specified. Furthermore, an image search viewer called Wisvi (Waseda Image Search VIewer) is developed and exploited. A β-version of all source codes can be down-loaded from a web site given in the text.

AB - New methods for joint compression and Image-to-Image retrieval (I2I retrieval) are presented. The novelty exists in the usage of computationally learned image bases besides color distributions. The bases are obtained by the Principal Component Analysis and/or the Independent Component Analysis. On the image compression, PCA and ICA outperform the JPEG's DCT. This superiority holds even if the bases and superposition coefficients are quantized and encoded. On the I2I retrieval, the precision-recall curve is used to measure the performance. It is found that adding the basis information always increases the baseline ability of the color information. Besides the retrieval evaluation, a unified image format called RIM (Retrieval-aware IMage format) for effective packing of codewords including bases is specified. Furthermore, an image search viewer called Wisvi (Waseda Image Search VIewer) is developed and exploited. A β-version of all source codes can be down-loaded from a web site given in the text.

UR - http://www.scopus.com/inward/record.url?scp=51749085590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51749085590&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2007.4371015

DO - 10.1109/IJCNN.2007.4371015

M3 - Conference contribution

AN - SCOPUS:51749085590

SN - 142441380X

SN - 9781424413805

SP - 546

EP - 551

BT - IEEE International Conference on Neural Networks - Conference Proceedings

T2 - 2007 International Joint Conference on Neural Networks, IJCNN 2007

Y2 - 12 August 2007 through 17 August 2007

ER -

Image-to-image retrieval using computationally learned bases and color information

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this