Learning feature relevance and similarity metrics in image databases

B. Bhanu, Jing Peng, Shan Qing

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

31 Citations (Scopus)

Abstract

Most of the current image retrieval systems use »one-shot» queries to a database to retrieve similar images. Typically a K-NN (nearest neighbor) kind of algorithm is used where the weights of the features that are used to represent images remain fixed (or manually tweaked by the user) in the computation of a given similarity metric. However, neither all of the features are equally important for a given query nor a similarity metric is optimal for all kinds of images in a database. The manual adjustment of these weights and the selection of similarity metric are exhausting. Moreover, they require a very sophisticated user. The authors present a novel image retrieval system that continuously learns the weights of features and selects an appropriate similarity metric based on the user's feedback given as positive or negative image examples. Experimental results are presented that provide the objective evaluation of learning behavior of the system for image retrieval.

Original languageEnglish
Title of host publicationProceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages14-18
Number of pages5
ISBN (Print)0818685441, 9780818685446
DOIs
StatePublished - 1 Jan 1998
Event1998 IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998 - Santa Barbara, United States
Duration: 21 Jun 1998 → …

Publication series

NameProceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998

Other

Other1998 IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998
CountryUnited States
CitySanta Barbara
Period21/06/98 → …

Fingerprint

Image retrieval
learning
learning behavior
Feedback
evaluation

Cite this

Bhanu, B., Peng, J., & Qing, S. (1998). Learning feature relevance and similarity metrics in image databases. In Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998 (pp. 14-18). [694471] (Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IVL.1998.694471
Bhanu, B. ; Peng, Jing ; Qing, Shan. / Learning feature relevance and similarity metrics in image databases. Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998. Institute of Electrical and Electronics Engineers Inc., 1998. pp. 14-18 (Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998).
@inproceedings{a2e7b4564bd1477f88e0f35dcdd1c21a,
title = "Learning feature relevance and similarity metrics in image databases",
abstract = "Most of the current image retrieval systems use »one-shot» queries to a database to retrieve similar images. Typically a K-NN (nearest neighbor) kind of algorithm is used where the weights of the features that are used to represent images remain fixed (or manually tweaked by the user) in the computation of a given similarity metric. However, neither all of the features are equally important for a given query nor a similarity metric is optimal for all kinds of images in a database. The manual adjustment of these weights and the selection of similarity metric are exhausting. Moreover, they require a very sophisticated user. The authors present a novel image retrieval system that continuously learns the weights of features and selects an appropriate similarity metric based on the user's feedback given as positive or negative image examples. Experimental results are presented that provide the objective evaluation of learning behavior of the system for image retrieval.",
author = "B. Bhanu and Jing Peng and Shan Qing",
year = "1998",
month = "1",
day = "1",
doi = "10.1109/IVL.1998.694471",
language = "English",
isbn = "0818685441",
series = "Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "14--18",
booktitle = "Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998",

}

Bhanu, B, Peng, J & Qing, S 1998, Learning feature relevance and similarity metrics in image databases. in Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998., 694471, Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998, Institute of Electrical and Electronics Engineers Inc., pp. 14-18, 1998 IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998, Santa Barbara, United States, 21/06/98. https://doi.org/10.1109/IVL.1998.694471

Learning feature relevance and similarity metrics in image databases. / Bhanu, B.; Peng, Jing; Qing, Shan.

Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998. Institute of Electrical and Electronics Engineers Inc., 1998. p. 14-18 694471 (Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998).

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

TY - GEN

T1 - Learning feature relevance and similarity metrics in image databases

AU - Bhanu, B.

AU - Peng, Jing

AU - Qing, Shan

PY - 1998/1/1

Y1 - 1998/1/1

N2 - Most of the current image retrieval systems use »one-shot» queries to a database to retrieve similar images. Typically a K-NN (nearest neighbor) kind of algorithm is used where the weights of the features that are used to represent images remain fixed (or manually tweaked by the user) in the computation of a given similarity metric. However, neither all of the features are equally important for a given query nor a similarity metric is optimal for all kinds of images in a database. The manual adjustment of these weights and the selection of similarity metric are exhausting. Moreover, they require a very sophisticated user. The authors present a novel image retrieval system that continuously learns the weights of features and selects an appropriate similarity metric based on the user's feedback given as positive or negative image examples. Experimental results are presented that provide the objective evaluation of learning behavior of the system for image retrieval.

AB - Most of the current image retrieval systems use »one-shot» queries to a database to retrieve similar images. Typically a K-NN (nearest neighbor) kind of algorithm is used where the weights of the features that are used to represent images remain fixed (or manually tweaked by the user) in the computation of a given similarity metric. However, neither all of the features are equally important for a given query nor a similarity metric is optimal for all kinds of images in a database. The manual adjustment of these weights and the selection of similarity metric are exhausting. Moreover, they require a very sophisticated user. The authors present a novel image retrieval system that continuously learns the weights of features and selects an appropriate similarity metric based on the user's feedback given as positive or negative image examples. Experimental results are presented that provide the objective evaluation of learning behavior of the system for image retrieval.

UR - http://www.scopus.com/inward/record.url?scp=84901404814&partnerID=8YFLogxK

U2 - 10.1109/IVL.1998.694471

DO - 10.1109/IVL.1998.694471

M3 - Conference contribution

AN - SCOPUS:84901404814

SN - 0818685441

SN - 9780818685446

T3 - Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998

SP - 14

EP - 18

BT - Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Bhanu B, Peng J, Qing S. Learning feature relevance and similarity metrics in image databases. In Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998. Institute of Electrical and Electronics Engineers Inc. 1998. p. 14-18. 694471. (Proceedings - IEEE Workshop on Content-Based Access of Image and Video Libraries, CBAIVL 1998). https://doi.org/10.1109/IVL.1998.694471