Multiple object retrieval for image databases using multiple instance learning and relevance feedback

Chengcui Zhang, Shu Ching Chen, Mei-Ling Shyu

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

14 Citations (Scopus)

Abstract

This paper proposes a method to effectively discover users' concept patterns when multiple objects of interests (e.g., foreground and background objects) are involved in content-based image retrieval. The proposed method incorporates Multiple Instance Learning into the user relevance feedback in a seamless way to discover where the user's most interested objects/regions and how to map the local features of that region(s) to user's high-level concepts. A three-layer neural network is used to model the underlying mapping progressively through the feedback and learning procedure.

Original languageEnglish
Title of host publication2004 IEEE International Conference on Multimedia and Expo (ICME)
Pages775-778
Number of pages4
Volume2
StatePublished - Dec 1 2004
Event2004 IEEE International Conference on Multimedia and Expo (ICME) - Taipei, Taiwan, Province of China
Duration: Jun 27 2004Jun 30 2004

Other

Other2004 IEEE International Conference on Multimedia and Expo (ICME)
CountryTaiwan, Province of China
CityTaipei
Period6/27/046/30/04

Fingerprint

Feedback
Image retrieval
Neural networks

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Zhang, C., Chen, S. C., & Shyu, M-L. (2004). Multiple object retrieval for image databases using multiple instance learning and relevance feedback. In 2004 IEEE International Conference on Multimedia and Expo (ICME) (Vol. 2, pp. 775-778)

Multiple object retrieval for image databases using multiple instance learning and relevance feedback. / Zhang, Chengcui; Chen, Shu Ching; Shyu, Mei-Ling.

2004 IEEE International Conference on Multimedia and Expo (ICME). Vol. 2 2004. p. 775-778.

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

Zhang, C, Chen, SC & Shyu, M-L 2004, Multiple object retrieval for image databases using multiple instance learning and relevance feedback. in 2004 IEEE International Conference on Multimedia and Expo (ICME). vol. 2, pp. 775-778, 2004 IEEE International Conference on Multimedia and Expo (ICME), Taipei, Taiwan, Province of China, 6/27/04.
Zhang C, Chen SC, Shyu M-L. Multiple object retrieval for image databases using multiple instance learning and relevance feedback. In 2004 IEEE International Conference on Multimedia and Expo (ICME). Vol. 2. 2004. p. 775-778
Zhang, Chengcui ; Chen, Shu Ching ; Shyu, Mei-Ling. / Multiple object retrieval for image databases using multiple instance learning and relevance feedback. 2004 IEEE International Conference on Multimedia and Expo (ICME). Vol. 2 2004. pp. 775-778
@inproceedings{a86e2475dfd146c585ea338a825b8348,
title = "Multiple object retrieval for image databases using multiple instance learning and relevance feedback",
abstract = "This paper proposes a method to effectively discover users' concept patterns when multiple objects of interests (e.g., foreground and background objects) are involved in content-based image retrieval. The proposed method incorporates Multiple Instance Learning into the user relevance feedback in a seamless way to discover where the user's most interested objects/regions and how to map the local features of that region(s) to user's high-level concepts. A three-layer neural network is used to model the underlying mapping progressively through the feedback and learning procedure.",
author = "Chengcui Zhang and Chen, {Shu Ching} and Mei-Ling Shyu",
year = "2004",
month = "12",
day = "1",
language = "English",
isbn = "0780386035",
volume = "2",
pages = "775--778",
booktitle = "2004 IEEE International Conference on Multimedia and Expo (ICME)",

}

TY - GEN

T1 - Multiple object retrieval for image databases using multiple instance learning and relevance feedback

AU - Zhang, Chengcui

AU - Chen, Shu Ching

AU - Shyu, Mei-Ling

PY - 2004/12/1

Y1 - 2004/12/1

N2 - This paper proposes a method to effectively discover users' concept patterns when multiple objects of interests (e.g., foreground and background objects) are involved in content-based image retrieval. The proposed method incorporates Multiple Instance Learning into the user relevance feedback in a seamless way to discover where the user's most interested objects/regions and how to map the local features of that region(s) to user's high-level concepts. A three-layer neural network is used to model the underlying mapping progressively through the feedback and learning procedure.

AB - This paper proposes a method to effectively discover users' concept patterns when multiple objects of interests (e.g., foreground and background objects) are involved in content-based image retrieval. The proposed method incorporates Multiple Instance Learning into the user relevance feedback in a seamless way to discover where the user's most interested objects/regions and how to map the local features of that region(s) to user's high-level concepts. A three-layer neural network is used to model the underlying mapping progressively through the feedback and learning procedure.

UR - http://www.scopus.com/inward/record.url?scp=11244255023&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=11244255023&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780386035

SN - 9780780386037

VL - 2

SP - 775

EP - 778

BT - 2004 IEEE International Conference on Multimedia and Expo (ICME)

ER -