An indexing scheme for efficient camera angle invariant image retrieval

Yoon S. Tak, Een Jun Hwang

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

4 Citations (Scopus)

Abstract

In specialized applications such as security, monitoring and surveillance, camera angle invariant image retrieval might be very useful. Typically, for shape-based angle invariant image retrieval, we need an object image at every angle where the shape of the image changes. This will generate a huge size of image database. Even worse, if we simply apply existing indexing schemes to this database, we will face serious performance problems due to the data size and data redundancy. In this paper, we propose a new indexing and matching scheme for shape-based angle invariant image retrieval, focusing on two types of camera movements: horizontal rotation and fixed rotation. To reduce the space requirement for indexing the huge database, we use two levels of indexing structures: (i) static indexing for horizontally rotated images for horizontal angle invariance, and (ii) dynamic indexing for fixedly rotated images for rotation invariance. Through the experiment, we show that our proposed scheme can achieve much better performance than other competing methods.

Original languageEnglish
Title of host publicationProceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008
Pages143-148
Number of pages6
DOIs
Publication statusPublished - 2008 Sep 22
Event2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008 - Sydney, NSW, Australia
Duration: 2008 Jul 82008 Jul 11

Other

Other2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008
CountryAustralia
CitySydney, NSW
Period08/7/808/7/11

Fingerprint

Image retrieval
Cameras
Invariance
Redundancy
Monitoring
Experiments

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems

Cite this

Tak, Y. S., & Hwang, E. J. (2008). An indexing scheme for efficient camera angle invariant image retrieval. In Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008 (pp. 143-148). [4594664] https://doi.org/10.1109/CIT.2008.4594664

An indexing scheme for efficient camera angle invariant image retrieval. / Tak, Yoon S.; Hwang, Een Jun.

Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008. 2008. p. 143-148 4594664.

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

Tak, YS & Hwang, EJ 2008, An indexing scheme for efficient camera angle invariant image retrieval. in Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008., 4594664, pp. 143-148, 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008, Sydney, NSW, Australia, 08/7/8. https://doi.org/10.1109/CIT.2008.4594664
Tak YS, Hwang EJ. An indexing scheme for efficient camera angle invariant image retrieval. In Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008. 2008. p. 143-148. 4594664 https://doi.org/10.1109/CIT.2008.4594664
Tak, Yoon S. ; Hwang, Een Jun. / An indexing scheme for efficient camera angle invariant image retrieval. Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008. 2008. pp. 143-148
@inproceedings{ea2086957a4c4fe19c9139325f62d39c,
title = "An indexing scheme for efficient camera angle invariant image retrieval",
abstract = "In specialized applications such as security, monitoring and surveillance, camera angle invariant image retrieval might be very useful. Typically, for shape-based angle invariant image retrieval, we need an object image at every angle where the shape of the image changes. This will generate a huge size of image database. Even worse, if we simply apply existing indexing schemes to this database, we will face serious performance problems due to the data size and data redundancy. In this paper, we propose a new indexing and matching scheme for shape-based angle invariant image retrieval, focusing on two types of camera movements: horizontal rotation and fixed rotation. To reduce the space requirement for indexing the huge database, we use two levels of indexing structures: (i) static indexing for horizontally rotated images for horizontal angle invariance, and (ii) dynamic indexing for fixedly rotated images for rotation invariance. Through the experiment, we show that our proposed scheme can achieve much better performance than other competing methods.",
author = "Tak, {Yoon S.} and Hwang, {Een Jun}",
year = "2008",
month = "9",
day = "22",
doi = "10.1109/CIT.2008.4594664",
language = "English",
isbn = "9781424423583",
pages = "143--148",
booktitle = "Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008",

}

TY - GEN

T1 - An indexing scheme for efficient camera angle invariant image retrieval

AU - Tak, Yoon S.

AU - Hwang, Een Jun

PY - 2008/9/22

Y1 - 2008/9/22

N2 - In specialized applications such as security, monitoring and surveillance, camera angle invariant image retrieval might be very useful. Typically, for shape-based angle invariant image retrieval, we need an object image at every angle where the shape of the image changes. This will generate a huge size of image database. Even worse, if we simply apply existing indexing schemes to this database, we will face serious performance problems due to the data size and data redundancy. In this paper, we propose a new indexing and matching scheme for shape-based angle invariant image retrieval, focusing on two types of camera movements: horizontal rotation and fixed rotation. To reduce the space requirement for indexing the huge database, we use two levels of indexing structures: (i) static indexing for horizontally rotated images for horizontal angle invariance, and (ii) dynamic indexing for fixedly rotated images for rotation invariance. Through the experiment, we show that our proposed scheme can achieve much better performance than other competing methods.

AB - In specialized applications such as security, monitoring and surveillance, camera angle invariant image retrieval might be very useful. Typically, for shape-based angle invariant image retrieval, we need an object image at every angle where the shape of the image changes. This will generate a huge size of image database. Even worse, if we simply apply existing indexing schemes to this database, we will face serious performance problems due to the data size and data redundancy. In this paper, we propose a new indexing and matching scheme for shape-based angle invariant image retrieval, focusing on two types of camera movements: horizontal rotation and fixed rotation. To reduce the space requirement for indexing the huge database, we use two levels of indexing structures: (i) static indexing for horizontally rotated images for horizontal angle invariance, and (ii) dynamic indexing for fixedly rotated images for rotation invariance. Through the experiment, we show that our proposed scheme can achieve much better performance than other competing methods.

UR - http://www.scopus.com/inward/record.url?scp=51849086390&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51849086390&partnerID=8YFLogxK

U2 - 10.1109/CIT.2008.4594664

DO - 10.1109/CIT.2008.4594664

M3 - Conference contribution

AN - SCOPUS:51849086390

SN - 9781424423583

SP - 143

EP - 148

BT - Proceedings - 2008 IEEE 8th International Conference on Computer and Information Technology, CIT 2008

ER -