Boosting web image search by co-ranking

Jingrui He, Changshui Zhang, Nanyuan Zhao, Hanghang Tong

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

5 Citations (Scopus)

Abstract

To maximally improve the precision among top-ranked images returned by a web image search engine without putting extra burden on the user, we propose in this paper a novel co-ranking framework which will re-rank the retrieved images to move the irrelevant ones to the tail of the list. The characteristic of the proposed framework can be summarized as follows: (1) making use of the decisions from multi-view of images to boost retrieval performance; (2) generalizing present multi-view algorithms which need labeled data for initialization to the unsupervised case so that no extra interaction is required. To implement the framework, we use one-class support vector machines to train the basic learner, and propose different schemes for combination. Experimental results demonstrate the effectiveness of the proposed framework.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
VolumeII
DOIs
StatePublished - 2005
Externally publishedYes
Event2005 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '05 - Philadelphia, PA, United States
Duration: Mar 18 2005Mar 23 2005

Other

Other2005 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '05
CountryUnited States
CityPhiladelphia, PA
Period3/18/053/23/05

Fingerprint

ranking
Search engines
World Wide Web
Support vector machines
acceleration (physics)
lists
retrieval
engines
interactions

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

He, J., Zhang, C., Zhao, N., & Tong, H. (2005). Boosting web image search by co-ranking. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (Vol. II). [1415428] https://doi.org/10.1109/ICASSP.2005.1415428

Boosting web image search by co-ranking. / He, Jingrui; Zhang, Changshui; Zhao, Nanyuan; Tong, Hanghang.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. II 2005. 1415428.

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

He, J, Zhang, C, Zhao, N & Tong, H 2005, Boosting web image search by co-ranking. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. vol. II, 1415428, 2005 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '05, Philadelphia, PA, United States, 3/18/05. https://doi.org/10.1109/ICASSP.2005.1415428
He J, Zhang C, Zhao N, Tong H. Boosting web image search by co-ranking. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. II. 2005. 1415428 https://doi.org/10.1109/ICASSP.2005.1415428
He, Jingrui ; Zhang, Changshui ; Zhao, Nanyuan ; Tong, Hanghang. / Boosting web image search by co-ranking. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. Vol. II 2005.
@inproceedings{e0a886225d3142b09d3ca0822971e928,
title = "Boosting web image search by co-ranking",
abstract = "To maximally improve the precision among top-ranked images returned by a web image search engine without putting extra burden on the user, we propose in this paper a novel co-ranking framework which will re-rank the retrieved images to move the irrelevant ones to the tail of the list. The characteristic of the proposed framework can be summarized as follows: (1) making use of the decisions from multi-view of images to boost retrieval performance; (2) generalizing present multi-view algorithms which need labeled data for initialization to the unsupervised case so that no extra interaction is required. To implement the framework, we use one-class support vector machines to train the basic learner, and propose different schemes for combination. Experimental results demonstrate the effectiveness of the proposed framework.",
author = "Jingrui He and Changshui Zhang and Nanyuan Zhao and Hanghang Tong",
year = "2005",
doi = "10.1109/ICASSP.2005.1415428",
language = "English (US)",
isbn = "0780388747",
volume = "II",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Boosting web image search by co-ranking

AU - He, Jingrui

AU - Zhang, Changshui

AU - Zhao, Nanyuan

AU - Tong, Hanghang

PY - 2005

Y1 - 2005

N2 - To maximally improve the precision among top-ranked images returned by a web image search engine without putting extra burden on the user, we propose in this paper a novel co-ranking framework which will re-rank the retrieved images to move the irrelevant ones to the tail of the list. The characteristic of the proposed framework can be summarized as follows: (1) making use of the decisions from multi-view of images to boost retrieval performance; (2) generalizing present multi-view algorithms which need labeled data for initialization to the unsupervised case so that no extra interaction is required. To implement the framework, we use one-class support vector machines to train the basic learner, and propose different schemes for combination. Experimental results demonstrate the effectiveness of the proposed framework.

AB - To maximally improve the precision among top-ranked images returned by a web image search engine without putting extra burden on the user, we propose in this paper a novel co-ranking framework which will re-rank the retrieved images to move the irrelevant ones to the tail of the list. The characteristic of the proposed framework can be summarized as follows: (1) making use of the decisions from multi-view of images to boost retrieval performance; (2) generalizing present multi-view algorithms which need labeled data for initialization to the unsupervised case so that no extra interaction is required. To implement the framework, we use one-class support vector machines to train the basic learner, and propose different schemes for combination. Experimental results demonstrate the effectiveness of the proposed framework.

UR - http://www.scopus.com/inward/record.url?scp=33646804891&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646804891&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2005.1415428

DO - 10.1109/ICASSP.2005.1415428

M3 - Conference contribution

SN - 0780388747

SN - 9780780388741

VL - II

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -