Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features

Bingrong Wang, Lei Sun, Jia Su, Takeshi Ikenaga

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper, we focus on the problem of complicated scene retrieval and give two proposals to improve accuracy of the recent image search system based on bag-of-features: block voting mechanism and weak feature selection. Both the methods aim to reduce effects of incorrect matching between descriptors. Block voting mechanism separates query and database images into blocks when computing image matching scores. It can be integrated into inverted file for an efficient and compact indexing structure. Weak feature selection provides a simple approach to select good feature points for matching. Experiments performed on a dataset with complicated scene and various transformations including viewpoint and illumination changes show an about 20 percent improvement rather than baseline bag-of-features due to my proposals.

Original languageEnglish
Title of host publicationNISS2010 - 4th International Conference on New Trends in Information Science and Service Science
Pages287-292
Number of pages6
Publication statusPublished - 2010
Event4th International Conference on New Trends in Information Science and Service Science, NISS2010 - Gyeongju
Duration: 2010 May 112010 May 13

Other

Other4th International Conference on New Trends in Information Science and Service Science, NISS2010
CityGyeongju
Period10/5/1110/5/13

Fingerprint

Feature selection
Voting
Integrated
Data base
Experiment
Inverted file
Query
Indexing

Keywords

  • Bag-of-features
  • Cotent-based image retrieval
  • Feature selection
  • Inverted file
  • Voting system

ASJC Scopus subject areas

  • Information Systems and Management

Cite this

Wang, B., Sun, L., Su, J., & Ikenaga, T. (2010). Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features. In NISS2010 - 4th International Conference on New Trends in Information Science and Service Science (pp. 287-292). [5488606]

Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features. / Wang, Bingrong; Sun, Lei; Su, Jia; Ikenaga, Takeshi.

NISS2010 - 4th International Conference on New Trends in Information Science and Service Science. 2010. p. 287-292 5488606.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wang, B, Sun, L, Su, J & Ikenaga, T 2010, Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features. in NISS2010 - 4th International Conference on New Trends in Information Science and Service Science., 5488606, pp. 287-292, 4th International Conference on New Trends in Information Science and Service Science, NISS2010, Gyeongju, 10/5/11.
Wang B, Sun L, Su J, Ikenaga T. Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features. In NISS2010 - 4th International Conference on New Trends in Information Science and Service Science. 2010. p. 287-292. 5488606
Wang, Bingrong ; Sun, Lei ; Su, Jia ; Ikenaga, Takeshi. / Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features. NISS2010 - 4th International Conference on New Trends in Information Science and Service Science. 2010. pp. 287-292
@inproceedings{7f9602229e9640c8af89affb5cb6e4d3,
title = "Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features",
abstract = "In this paper, we focus on the problem of complicated scene retrieval and give two proposals to improve accuracy of the recent image search system based on bag-of-features: block voting mechanism and weak feature selection. Both the methods aim to reduce effects of incorrect matching between descriptors. Block voting mechanism separates query and database images into blocks when computing image matching scores. It can be integrated into inverted file for an efficient and compact indexing structure. Weak feature selection provides a simple approach to select good feature points for matching. Experiments performed on a dataset with complicated scene and various transformations including viewpoint and illumination changes show an about 20 percent improvement rather than baseline bag-of-features due to my proposals.",
keywords = "Bag-of-features, Cotent-based image retrieval, Feature selection, Inverted file, Voting system",
author = "Bingrong Wang and Lei Sun and Jia Su and Takeshi Ikenaga",
year = "2010",
language = "English",
isbn = "9788988678183",
pages = "287--292",
booktitle = "NISS2010 - 4th International Conference on New Trends in Information Science and Service Science",

}

TY - GEN

T1 - Complicated scene retrieval using block voting mechanism and weak feature selection based on bag-of-features

AU - Wang, Bingrong

AU - Sun, Lei

AU - Su, Jia

AU - Ikenaga, Takeshi

PY - 2010

Y1 - 2010

N2 - In this paper, we focus on the problem of complicated scene retrieval and give two proposals to improve accuracy of the recent image search system based on bag-of-features: block voting mechanism and weak feature selection. Both the methods aim to reduce effects of incorrect matching between descriptors. Block voting mechanism separates query and database images into blocks when computing image matching scores. It can be integrated into inverted file for an efficient and compact indexing structure. Weak feature selection provides a simple approach to select good feature points for matching. Experiments performed on a dataset with complicated scene and various transformations including viewpoint and illumination changes show an about 20 percent improvement rather than baseline bag-of-features due to my proposals.

AB - In this paper, we focus on the problem of complicated scene retrieval and give two proposals to improve accuracy of the recent image search system based on bag-of-features: block voting mechanism and weak feature selection. Both the methods aim to reduce effects of incorrect matching between descriptors. Block voting mechanism separates query and database images into blocks when computing image matching scores. It can be integrated into inverted file for an efficient and compact indexing structure. Weak feature selection provides a simple approach to select good feature points for matching. Experiments performed on a dataset with complicated scene and various transformations including viewpoint and illumination changes show an about 20 percent improvement rather than baseline bag-of-features due to my proposals.

KW - Bag-of-features

KW - Cotent-based image retrieval

KW - Feature selection

KW - Inverted file

KW - Voting system

UR - http://www.scopus.com/inward/record.url?scp=77957766932&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957766932&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:77957766932

SN - 9788988678183

SP - 287

EP - 292

BT - NISS2010 - 4th International Conference on New Trends in Information Science and Service Science

ER -