Learning a cross-modal hashing network for multimedia search
In this paper, we propose a cross-modal hashing network (CMHN) method to learn compact binary codes for cross-modality multimedia search. Unlike most existing cross-modal hashing methods which learn a single pair of projections to map each example into a binary vector, we design a deep neural networ...
Saved in:
Main Authors: | , , |
---|---|
Other Authors: | |
Format: | Conference or Workshop Item |
Language: | English |
Published: |
2018
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/85331 http://hdl.handle.net/10220/44604 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-85331 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-853312020-11-01T04:43:03Z Learning a cross-modal hashing network for multimedia search Tan, Yap Peng Liong, Venice Erin Lu, Jiwen School of Electrical and Electronic Engineering Interdisciplinary Graduate School (IGS) 2017 IEEE International Conference on Image Processing (ICIP) Hashing Cross-modal Retrieval In this paper, we propose a cross-modal hashing network (CMHN) method to learn compact binary codes for cross-modality multimedia search. Unlike most existing cross-modal hashing methods which learn a single pair of projections to map each example into a binary vector, we design a deep neural network to learn multiple pairs of hierarchical non-linear transformations, under which the nonlinear characteristics of samples can be well exploited and the modality gap is well reduced. Our model is trained under an iterative optimization procedure which learns a (1) unified binary code discretely and discriminatively through a classification-based hinge-loss criterion, and (2) cross-modal hashing network, one deep network for each modality, through minimizing the quantization loss between real-valued neural code and binary code, and maximizing the variance of the learned neural codes. Experimental results on two benchmark datasets show the efficacy of the proposed approach. Accepted version 2018-03-23T06:16:21Z 2019-12-06T16:01:45Z 2018-03-23T06:16:21Z 2019-12-06T16:01:45Z 2017 Conference Paper Liong, V. E., Lu, J., & Tan, Y.-P. (2017, September). Learning a cross-modal hashing network for multimedia search. Paper presented at 2017 IEEE International Conference on Image Processing (ICIP), Beijing, China (pp. 3700-3704). IEEE. https://hdl.handle.net/10356/85331 http://hdl.handle.net/10220/44604 10.1109/ICIP.2017.8296973 en © 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: [http://dx.doi.org/10.1109/ICIP.2017.8296973]. 5 p. application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Hashing Cross-modal Retrieval |
spellingShingle |
Hashing Cross-modal Retrieval Tan, Yap Peng Liong, Venice Erin Lu, Jiwen Learning a cross-modal hashing network for multimedia search |
description |
In this paper, we propose a cross-modal hashing network (CMHN) method to learn compact binary codes for cross-modality multimedia search. Unlike most existing cross-modal hashing methods which learn a single pair of projections to map each example into a binary vector, we design a deep neural network to learn multiple pairs of hierarchical non-linear transformations, under which the nonlinear characteristics of samples can be well exploited and the modality gap is well reduced. Our model is trained under an iterative optimization procedure which learns a (1) unified binary code discretely and discriminatively through a classification-based hinge-loss criterion, and (2) cross-modal hashing network, one deep network for each modality, through minimizing the quantization loss between real-valued neural code and binary code, and maximizing the variance of the learned neural codes. Experimental results on two benchmark datasets show the efficacy of the proposed approach. |
author2 |
School of Electrical and Electronic Engineering |
author_facet |
School of Electrical and Electronic Engineering Tan, Yap Peng Liong, Venice Erin Lu, Jiwen |
format |
Conference or Workshop Item |
author |
Tan, Yap Peng Liong, Venice Erin Lu, Jiwen |
author_sort |
Tan, Yap Peng |
title |
Learning a cross-modal hashing network for multimedia search |
title_short |
Learning a cross-modal hashing network for multimedia search |
title_full |
Learning a cross-modal hashing network for multimedia search |
title_fullStr |
Learning a cross-modal hashing network for multimedia search |
title_full_unstemmed |
Learning a cross-modal hashing network for multimedia search |
title_sort |
learning a cross-modal hashing network for multimedia search |
publishDate |
2018 |
url |
https://hdl.handle.net/10356/85331 http://hdl.handle.net/10220/44604 |
_version_ |
1683493137333354496 |