Optimizing progressive query-by-example over pre-clustered large image databases

Anicet Kouomou Choupo, Laure Berti-Equille, Annie Morin

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

3 Citations (Scopus)

Abstract

The typical mode for querying in an image content-based information system is query-by-example, which allows the user to provide an image as a query and to search for similar images (i.e., the nearest neighbors) based on one or a combination of low-level multidimensional features of the query example. Off-lime, this requires the time-consuming pre-computing of the whole set of visual descriptors over the image database. On-line, one major drawback is that multidimensional sequential NN-search is usually exhaustive over the whole image set face to the user who has a very limited patience. In this paper, we propose a technique for improving the performance of image query-by-example execution strategies over multiple visual features. This includes first, the pre-clustering of the large image database and then, the scheduling of the processing of the feature clusters before providing progressively the query results (i.e., intermediate results are sent continuously before the end of the exhaustive scan over the whole database). A cluster eligibility criterion and two filtering rules are proposed to select the most relevant clusters to a query-by-example. Experiments over more than 110 000 images and five MPEG-7 global features show that our approach significantly reduces the query time in two experimental cases: the query time is divided by 4.8 for 100 clusters per descriptor type and by 7 for 200 clusters per descriptor type compared to a "blind" sequential NN-search with keeping the same final query result. This constitutes a promising perspective for optimizing image query-by-example execution.

Original languageEnglish
Title of host publicationACM International Conference Proceeding Series
Pages13-20
Number of pages8
Volume160
DOIs
Publication statusPublished - 1 Dec 2005
Externally publishedYes
Event2nd International Workshop on Computer Vision Meets Databases, CVDB '05 - Baltimore, MD, United States
Duration: 17 Jun 200517 Jun 2005

Other

Other2nd International Workshop on Computer Vision Meets Databases, CVDB '05
CountryUnited States
CityBaltimore, MD
Period17/6/0517/6/05

Fingerprint

Lime
Information systems
Scheduling
Processing
Experiments

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Choupo, A. K., Berti-Equille, L., & Morin, A. (2005). Optimizing progressive query-by-example over pre-clustered large image databases. In ACM International Conference Proceeding Series (Vol. 160, pp. 13-20) https://doi.org/10.1145/1160939.1160946

Optimizing progressive query-by-example over pre-clustered large image databases. / Choupo, Anicet Kouomou; Berti-Equille, Laure; Morin, Annie.

ACM International Conference Proceeding Series. Vol. 160 2005. p. 13-20.

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

Choupo, AK, Berti-Equille, L & Morin, A 2005, Optimizing progressive query-by-example over pre-clustered large image databases. in ACM International Conference Proceeding Series. vol. 160, pp. 13-20, 2nd International Workshop on Computer Vision Meets Databases, CVDB '05, Baltimore, MD, United States, 17/6/05. https://doi.org/10.1145/1160939.1160946
Choupo AK, Berti-Equille L, Morin A. Optimizing progressive query-by-example over pre-clustered large image databases. In ACM International Conference Proceeding Series. Vol. 160. 2005. p. 13-20 https://doi.org/10.1145/1160939.1160946
Choupo, Anicet Kouomou ; Berti-Equille, Laure ; Morin, Annie. / Optimizing progressive query-by-example over pre-clustered large image databases. ACM International Conference Proceeding Series. Vol. 160 2005. pp. 13-20
@inproceedings{48ba5767fa4d4e30ac4f80ad97c1c122,
title = "Optimizing progressive query-by-example over pre-clustered large image databases",
abstract = "The typical mode for querying in an image content-based information system is query-by-example, which allows the user to provide an image as a query and to search for similar images (i.e., the nearest neighbors) based on one or a combination of low-level multidimensional features of the query example. Off-lime, this requires the time-consuming pre-computing of the whole set of visual descriptors over the image database. On-line, one major drawback is that multidimensional sequential NN-search is usually exhaustive over the whole image set face to the user who has a very limited patience. In this paper, we propose a technique for improving the performance of image query-by-example execution strategies over multiple visual features. This includes first, the pre-clustering of the large image database and then, the scheduling of the processing of the feature clusters before providing progressively the query results (i.e., intermediate results are sent continuously before the end of the exhaustive scan over the whole database). A cluster eligibility criterion and two filtering rules are proposed to select the most relevant clusters to a query-by-example. Experiments over more than 110 000 images and five MPEG-7 global features show that our approach significantly reduces the query time in two experimental cases: the query time is divided by 4.8 for 100 clusters per descriptor type and by 7 for 200 clusters per descriptor type compared to a {"}blind{"} sequential NN-search with keeping the same final query result. This constitutes a promising perspective for optimizing image query-by-example execution.",
author = "Choupo, {Anicet Kouomou} and Laure Berti-Equille and Annie Morin",
year = "2005",
month = "12",
day = "1",
doi = "10.1145/1160939.1160946",
language = "English",
isbn = "1595931511",
volume = "160",
pages = "13--20",
booktitle = "ACM International Conference Proceeding Series",

}

TY - GEN

T1 - Optimizing progressive query-by-example over pre-clustered large image databases

AU - Choupo, Anicet Kouomou

AU - Berti-Equille, Laure

AU - Morin, Annie

PY - 2005/12/1

Y1 - 2005/12/1

N2 - The typical mode for querying in an image content-based information system is query-by-example, which allows the user to provide an image as a query and to search for similar images (i.e., the nearest neighbors) based on one or a combination of low-level multidimensional features of the query example. Off-lime, this requires the time-consuming pre-computing of the whole set of visual descriptors over the image database. On-line, one major drawback is that multidimensional sequential NN-search is usually exhaustive over the whole image set face to the user who has a very limited patience. In this paper, we propose a technique for improving the performance of image query-by-example execution strategies over multiple visual features. This includes first, the pre-clustering of the large image database and then, the scheduling of the processing of the feature clusters before providing progressively the query results (i.e., intermediate results are sent continuously before the end of the exhaustive scan over the whole database). A cluster eligibility criterion and two filtering rules are proposed to select the most relevant clusters to a query-by-example. Experiments over more than 110 000 images and five MPEG-7 global features show that our approach significantly reduces the query time in two experimental cases: the query time is divided by 4.8 for 100 clusters per descriptor type and by 7 for 200 clusters per descriptor type compared to a "blind" sequential NN-search with keeping the same final query result. This constitutes a promising perspective for optimizing image query-by-example execution.

AB - The typical mode for querying in an image content-based information system is query-by-example, which allows the user to provide an image as a query and to search for similar images (i.e., the nearest neighbors) based on one or a combination of low-level multidimensional features of the query example. Off-lime, this requires the time-consuming pre-computing of the whole set of visual descriptors over the image database. On-line, one major drawback is that multidimensional sequential NN-search is usually exhaustive over the whole image set face to the user who has a very limited patience. In this paper, we propose a technique for improving the performance of image query-by-example execution strategies over multiple visual features. This includes first, the pre-clustering of the large image database and then, the scheduling of the processing of the feature clusters before providing progressively the query results (i.e., intermediate results are sent continuously before the end of the exhaustive scan over the whole database). A cluster eligibility criterion and two filtering rules are proposed to select the most relevant clusters to a query-by-example. Experiments over more than 110 000 images and five MPEG-7 global features show that our approach significantly reduces the query time in two experimental cases: the query time is divided by 4.8 for 100 clusters per descriptor type and by 7 for 200 clusters per descriptor type compared to a "blind" sequential NN-search with keeping the same final query result. This constitutes a promising perspective for optimizing image query-by-example execution.

UR - http://www.scopus.com/inward/record.url?scp=77953517982&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953517982&partnerID=8YFLogxK

U2 - 10.1145/1160939.1160946

DO - 10.1145/1160939.1160946

M3 - Conference contribution

SN - 1595931511

SN - 9781595931511

VL - 160

SP - 13

EP - 20

BT - ACM International Conference Proceeding Series

ER -