Incremental hashing with undersampling

Xiaoxia Jiang; Wing W.Y. Ng; Xing Tian; Sam Kwong; Hui Wang

doi:10.1109/SMC.2019.8914271

Incremental hashing with undersampling

Xiaoxia Jiang, Wing W.Y. Ng^*, Xing Tian, Sam Kwong, Hui Wang

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

Most of current hashing methods are proposed based on the assumption that the database is stationary. However, this assumption is not always true as the data environment is sometimes non-stationary. When new images being added to the database, data distributions of existing classes may change and new classes may also appear which result in concept drifts. The problem of concept drifts is unavoidable in non-stationary data environments. Incremental Hashing (ICH) is an effective method for image retrieval in non-stationary data environments with concept drifts using multiple hash tables. In ICH, new concept is adapted by training new hash table using the most updated data chunks. However, images in the new data chunk may not be all informative for updating. To enhance the efficiency of ICH, ICH with Undersampling (ICHUS) is proposed to select informative samples in the new data chunk for the training of new hash table to adapt to the non-stationary data environment. Experimental results show that ICHUS yields a better retrieval performance than ICH and state-of-art non-stationary hashing methods.

Original language	English
Title of host publication	2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	616-622
Number of pages	7
ISBN (Electronic)	9781728145693
ISBN (Print)	9781728145709
DOIs	https://doi.org/10.1109/SMC.2019.8914271
Publication status	Published - 28 Nov 2019
Externally published	Yes
Event	2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019 - Bari, Italy Duration: 06 Oct 2019 → 09 Oct 2019

Publication series

Name	Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics
Volume	2019-October
ISSN (Print)	1062-922X

Conference

Conference	2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019
Country/Territory	Italy
City	Bari
Period	06/10/2019 → 09/10/2019

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Keywords

Concept Drifts
Image Retrieval
Incremental Hashing
Semi-supervised Hashing
Undersampling

ASJC Scopus subject areas

Electrical and Electronic Engineering
Control and Systems Engineering
Human-Computer Interaction

Access to Document

10.1109/SMC.2019.8914271Licence: Unspecified

Cite this

Jiang, X., Ng, W. W. Y., Tian, X., Kwong, S., & Wang, H. (2019). Incremental hashing with undersampling. In 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings (pp. 616-622). Article 8914271 (Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics; Vol. 2019-October). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SMC.2019.8914271

@inproceedings{b5f1bf5d16c24f58b1fc75c3f290708b,

title = "Incremental hashing with undersampling",

abstract = "Most of current hashing methods are proposed based on the assumption that the database is stationary. However, this assumption is not always true as the data environment is sometimes non-stationary. When new images being added to the database, data distributions of existing classes may change and new classes may also appear which result in concept drifts. The problem of concept drifts is unavoidable in non-stationary data environments. Incremental Hashing (ICH) is an effective method for image retrieval in non-stationary data environments with concept drifts using multiple hash tables. In ICH, new concept is adapted by training new hash table using the most updated data chunks. However, images in the new data chunk may not be all informative for updating. To enhance the efficiency of ICH, ICH with Undersampling (ICHUS) is proposed to select informative samples in the new data chunk for the training of new hash table to adapt to the non-stationary data environment. Experimental results show that ICHUS yields a better retrieval performance than ICH and state-of-art non-stationary hashing methods.",

keywords = "Concept Drifts, Image Retrieval, Incremental Hashing, Semi-supervised Hashing, Undersampling",

author = "Xiaoxia Jiang and Ng, {Wing W.Y.} and Xing Tian and Sam Kwong and Hui Wang",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019 ; Conference date: 06-10-2019 Through 09-10-2019",

year = "2019",

month = nov,

day = "28",

doi = "10.1109/SMC.2019.8914271",

language = "English",

isbn = "9781728145709",

series = "Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "616--622",

booktitle = "2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings",

address = "United States",

}

Jiang, X, Ng, WWY, Tian, X, Kwong, S & Wang, H 2019, Incremental hashing with undersampling. in 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings., 8914271, Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics, vol. 2019-October, Institute of Electrical and Electronics Engineers Inc., pp. 616-622, 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019, Bari, Italy, 06/10/2019. https://doi.org/10.1109/SMC.2019.8914271

Incremental hashing with undersampling. / Jiang, Xiaoxia; Ng, Wing W.Y.; Tian, Xing et al.
2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 616-622 8914271 (Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics; Vol. 2019-October).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Incremental hashing with undersampling

AU - Jiang, Xiaoxia

AU - Ng, Wing W.Y.

AU - Tian, Xing

AU - Kwong, Sam

AU - Wang, Hui

PY - 2019/11/28

Y1 - 2019/11/28

N2 - Most of current hashing methods are proposed based on the assumption that the database is stationary. However, this assumption is not always true as the data environment is sometimes non-stationary. When new images being added to the database, data distributions of existing classes may change and new classes may also appear which result in concept drifts. The problem of concept drifts is unavoidable in non-stationary data environments. Incremental Hashing (ICH) is an effective method for image retrieval in non-stationary data environments with concept drifts using multiple hash tables. In ICH, new concept is adapted by training new hash table using the most updated data chunks. However, images in the new data chunk may not be all informative for updating. To enhance the efficiency of ICH, ICH with Undersampling (ICHUS) is proposed to select informative samples in the new data chunk for the training of new hash table to adapt to the non-stationary data environment. Experimental results show that ICHUS yields a better retrieval performance than ICH and state-of-art non-stationary hashing methods.

AB - Most of current hashing methods are proposed based on the assumption that the database is stationary. However, this assumption is not always true as the data environment is sometimes non-stationary. When new images being added to the database, data distributions of existing classes may change and new classes may also appear which result in concept drifts. The problem of concept drifts is unavoidable in non-stationary data environments. Incremental Hashing (ICH) is an effective method for image retrieval in non-stationary data environments with concept drifts using multiple hash tables. In ICH, new concept is adapted by training new hash table using the most updated data chunks. However, images in the new data chunk may not be all informative for updating. To enhance the efficiency of ICH, ICH with Undersampling (ICHUS) is proposed to select informative samples in the new data chunk for the training of new hash table to adapt to the non-stationary data environment. Experimental results show that ICHUS yields a better retrieval performance than ICH and state-of-art non-stationary hashing methods.

KW - Concept Drifts

KW - Image Retrieval

KW - Incremental Hashing

KW - Semi-supervised Hashing

KW - Undersampling

U2 - 10.1109/SMC.2019.8914271

DO - 10.1109/SMC.2019.8914271

M3 - Conference contribution

AN - SCOPUS:85076790938

SN - 9781728145709

T3 - Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics

SP - 616

EP - 622

BT - 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019

Y2 - 6 October 2019 through 9 October 2019

ER -

Jiang X, Ng WWY, Tian X, Kwong S, Wang H. Incremental hashing with undersampling. In 2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019: proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 616-622. 8914271. (Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics). doi: 10.1109/SMC.2019.8914271

Incremental hashing with undersampling

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this