[1] Xu Jie, Shi Pengfei,. Active learning based on maximizing information gainfor content-based image retrieval [J]. Journal of Southeast University (English Edition), 2004, 20 (4): 431-435. [doi:10.3969/j.issn.1003-7985.2004.04.008]

Copy

Active learning based on maximizing information gainfor content-based image retrieval()

Share：

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:

20

Issue:

2004 4

Page:

431-435

Research Field:

Computer Science and Engineering

Publishing date:

2004-12-30

Info

Title:

Active learning based on maximizing information gainfor content-based image retrieval

Author(s):

Xu Jie;  Shi Pengfei

School of Electronic Information and Electrical Engineering, Shanghai Jiaotong University, Shanghai 200030, China

Keywords:

active learning;  content-based image retrieval;  relevance feedback;  support vector machines;  similarity measure

PACS:

TP391

DOI:

10.3969/j.issn.1003-7985.2004.04.008

Abstract:

This paper describes a new method for active learning in content-based image retrieval. The proposed method firstly uses support vector machine(SVM)classifiers to learn an initial query concept. Then the proposed active learning scheme employs similarity measure to check the current version space and selects images with maximum expected information gain to solicit user’s label. Finally, the learned query is refined based on the user’s further feedback. With the combination of SVM classifier and similarity measure, the proposed method can alleviate model bias existing in each of them. Our experiments on several query concepts show that the proposed method can learn the user’s query concept quickly and effectively only with several iterations.

References:

[1] Rui Y, Huang T S, Ortega M, et al. Relevance feedback: a power tool in interactive content-based image retrieval[J]. IEEE Trans Circuits and Systems for Video Tech, 1998, 8(5): 644-655.
[2] Seung H S, Opper M, Sompolinsky H. Query by committee[A]. In: Proc of the Fifth Workshop on Computational Learning Theory [C]. San Mateo, CA, 1992. 287-294.
[3] Freund Y, Seung H, Shamir E, et al. Selective sampling using the query by committee algorithm [J]. Machine Learning, 1997, 28(2, 3): 133-168.
[4] Chang E, Li B. MEGA — the maximizing expected generalization algorithm for learning complex query concepts [J]. ACM Transactions on Information Systems, 2003, 21(4): 347-382.
[5] Dagan I, Engelson S. Committee-based sampling for training probabilistic classifiers[A]. In: Proc of the Twelfth International Conference on Machine Learning [C]. Tahoe City, California, 1995. 150-157.
[6] Cohn D, Atlas L, Ladner R. Improving generalization with active learning [J]. Machine Learning, 1994, 15(2): 201-221.
[7] Lewis D, Gale W. A sequential algorithm for training text classifiers [A]. In: Proc of the Seventeenth Annual International ACM-SIGIR Conference on Research and Development in Information Retrieval [C]. Dublin, Ireland, 1994. 3-12.
[8] Tong S, Chang E. Support vector machine active learning for image retrieval [A]. In: Proc of ACM Mutimedia [C]. Ottawa, Canada, 2001. 107-118.
[9] Campbell C, Cristianini N, Smola A. Query learning with large margin classifiers [A]. In: Proc of the Seventeenth International Conference on Machine Learning[C]. Stanford University, 2000. 111-118.
[10] Schohn G, Cohn D. Less is more: active learning with support vector machines [A]. In: Proc of the Seventeenth International Conference on Machine Learning [C]. Stanford University, 2000. 839-846.
[11] Vapnik V. Statistical learning theory[M]. New York: Addison Wiley, 1998.
[12] Carson C, Belongie S, Greenspan H.et al. Blobworld: image segmentation using expectation-maximization and its application to image querying[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2002, 24(8): 1026-1038.

Memo

Memo:

Biographies: Xu Jie(1975—), female, graduate; Shi Pengfei(corresponding author), male, professor, pfshi@sjtu.edu.cn.

Common functions