Home > Research > Publications & Outputs > Cross-View Retrieval via Probability-Based Sema...

Associated organisational unit

Electronic data

  • double

    Rights statement: ©2017 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

    Accepted author manuscript, 1.03 MB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing. / Lin, Zijia; Ding, Guiguang; Han, Jungong et al.
In: IEEE Transactions on Cybernetics, Vol. 47, No. 12, 20.12.2017, p. 4342-4355.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Lin, Z, Ding, G, Han, J & Wang, J 2017, 'Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing', IEEE Transactions on Cybernetics, vol. 47, no. 12, pp. 4342-4355. https://doi.org/10.1109/TCYB.2016.2608906

APA

Lin, Z., Ding, G., Han, J., & Wang, J. (2017). Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing. IEEE Transactions on Cybernetics, 47(12), 4342-4355. https://doi.org/10.1109/TCYB.2016.2608906

Vancouver

Lin Z, Ding G, Han J, Wang J. Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing. IEEE Transactions on Cybernetics. 2017 Dec 20;47(12):4342-4355. Epub 2016 Sept 29. doi: 10.1109/TCYB.2016.2608906

Author

Lin, Zijia ; Ding, Guiguang ; Han, Jungong et al. / Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing. In: IEEE Transactions on Cybernetics. 2017 ; Vol. 47, No. 12. pp. 4342-4355.

Bibtex

@article{cf90a1c09b224fffbd41d0332d27bd6b,
title = "Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing",
abstract = "For efficiently retrieving nearest neighbors from large-scale multiview data, recently hashing methods are widely investigated, which can substantially improve query speeds. In this paper, we propose an effective probability-based semantics-preserving hashing (SePH) method to tackle the problem of cross-view retrieval. Considering the semantic consistency between views, SePH generates one unified hash code for all observed views of any instance. For training, SePH first transforms the given semantic affinities of training data into a probability distribution, and aims to approximate it with another one in Hamming space, via minimizing their Kullback-Leibler divergence. Specifically, the latter probability distribution is derived from all pair-wise Hamming distances between to-be-learnt hash codes of the training data. Then with learnt hash codes, any kind of predictive models like linear ridge regression, logistic regression, or kernel logistic regression, can be learnt as hash functions in each view for projecting the corresponding view-specific features into hash codes. As for out-of-sample extension, given any unseen instance, the learnt hash functions in its observed views can predict view-specific hash codes. Then by deriving or estimating the corresponding output probabilities with respect to the predicted view-specific hash codes, a novel probabilistic approach is further proposed to utilize them for determining a unified hash code. To evaluate the proposed SePH, we conduct extensive experiments on diverse benchmark datasets, and the experimental results demonstrate that SePH is reasonable and effective.",
author = "Zijia Lin and Guiguang Ding and Jungong Han and Jianmin Wang",
year = "2017",
month = dec,
day = "20",
doi = "10.1109/TCYB.2016.2608906",
language = "English",
volume = "47",
pages = "4342--4355",
journal = "IEEE Transactions on Cybernetics",
issn = "2168-2267",
publisher = "IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC",
number = "12",

}

RIS

TY - JOUR

T1 - Cross-View Retrieval via Probability-Based Semantics-Preserving Hashing

AU - Lin, Zijia

AU - Ding, Guiguang

AU - Han, Jungong

AU - Wang, Jianmin

PY - 2017/12/20

Y1 - 2017/12/20

N2 - For efficiently retrieving nearest neighbors from large-scale multiview data, recently hashing methods are widely investigated, which can substantially improve query speeds. In this paper, we propose an effective probability-based semantics-preserving hashing (SePH) method to tackle the problem of cross-view retrieval. Considering the semantic consistency between views, SePH generates one unified hash code for all observed views of any instance. For training, SePH first transforms the given semantic affinities of training data into a probability distribution, and aims to approximate it with another one in Hamming space, via minimizing their Kullback-Leibler divergence. Specifically, the latter probability distribution is derived from all pair-wise Hamming distances between to-be-learnt hash codes of the training data. Then with learnt hash codes, any kind of predictive models like linear ridge regression, logistic regression, or kernel logistic regression, can be learnt as hash functions in each view for projecting the corresponding view-specific features into hash codes. As for out-of-sample extension, given any unseen instance, the learnt hash functions in its observed views can predict view-specific hash codes. Then by deriving or estimating the corresponding output probabilities with respect to the predicted view-specific hash codes, a novel probabilistic approach is further proposed to utilize them for determining a unified hash code. To evaluate the proposed SePH, we conduct extensive experiments on diverse benchmark datasets, and the experimental results demonstrate that SePH is reasonable and effective.

AB - For efficiently retrieving nearest neighbors from large-scale multiview data, recently hashing methods are widely investigated, which can substantially improve query speeds. In this paper, we propose an effective probability-based semantics-preserving hashing (SePH) method to tackle the problem of cross-view retrieval. Considering the semantic consistency between views, SePH generates one unified hash code for all observed views of any instance. For training, SePH first transforms the given semantic affinities of training data into a probability distribution, and aims to approximate it with another one in Hamming space, via minimizing their Kullback-Leibler divergence. Specifically, the latter probability distribution is derived from all pair-wise Hamming distances between to-be-learnt hash codes of the training data. Then with learnt hash codes, any kind of predictive models like linear ridge regression, logistic regression, or kernel logistic regression, can be learnt as hash functions in each view for projecting the corresponding view-specific features into hash codes. As for out-of-sample extension, given any unseen instance, the learnt hash functions in its observed views can predict view-specific hash codes. Then by deriving or estimating the corresponding output probabilities with respect to the predicted view-specific hash codes, a novel probabilistic approach is further proposed to utilize them for determining a unified hash code. To evaluate the proposed SePH, we conduct extensive experiments on diverse benchmark datasets, and the experimental results demonstrate that SePH is reasonable and effective.

U2 - 10.1109/TCYB.2016.2608906

DO - 10.1109/TCYB.2016.2608906

M3 - Journal article

VL - 47

SP - 4342

EP - 4355

JO - IEEE Transactions on Cybernetics

JF - IEEE Transactions on Cybernetics

SN - 2168-2267

IS - 12

ER -