Learning robust multi-label hashing for efficient image retrieval

Supervised hashing generally achieves superior performance over unsupervised or semi-supervised approaches by leveraging semantic labels. However, most existing supervised hashing techniques only deal with image samples with single label. Few of them properly address the practical problem concerning...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلفون الرئيسيون: CHEN, Haibao, ZHAO, Yuyan, ZHU, Lei, CHEN, Guilin, SUN, Kaichuan
التنسيق: text
اللغة:English
منشور في: Institutional Knowledge at Singapore Management University 2016
الموضوعات:
الوصول للمادة أونلاين:https://ink.library.smu.edu.sg/sis_research/3593
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
المؤسسة: Singapore Management University
اللغة: English
id sg-smu-ink.sis_research-4594
record_format dspace
spelling sg-smu-ink.sis_research-45942017-04-10T02:12:07Z Learning robust multi-label hashing for efficient image retrieval CHEN, Haibao ZHAO, Yuyan ZHU, Lei CHEN, Guilin SUN, Kaichuan Supervised hashing generally achieves superior performance over unsupervised or semi-supervised approaches by leveraging semantic labels. However, most existing supervised hashing techniques only deal with image samples with single label. Few of them properly address the practical problem concerning images with multiple labels, which is very common in real applications. In this paper, we seek to address the limitations of the existing schemes by proposing a novel approach, dubbed as Robust Multi-Label Hashing (RMLH). A label hypergraph is constructed to effectively capture high-order semantic correlations of images. And they are preserved into hashing codes with hypergraph consistency and direct label-hashing correlation. Besides, we impose a nuclear norm regularization on correlation matrix to maintain label correlations and robustly accommodate missing labels. Furthermore, an efficient algorithm based on Alternate Direction Method of Multipliers (ADMM) is developed to calculate the optimal hashing codes. Experiments demonstrate that RMLH can outperform state-of-the-art schemes and enjoy much better robustness against missing labels. 2016-01-01T08:00:00Z text https://ink.library.smu.edu.sg/sis_research/3593 info:doi/10.1007/978-3-319-48896-7_28 Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Correlation matrix Label hypergraph Robust multi-label hashing Databases and Information Systems Systems Architecture
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Correlation matrix
Label hypergraph
Robust multi-label hashing
Databases and Information Systems
Systems Architecture
spellingShingle Correlation matrix
Label hypergraph
Robust multi-label hashing
Databases and Information Systems
Systems Architecture
CHEN, Haibao
ZHAO, Yuyan
ZHU, Lei
CHEN, Guilin
SUN, Kaichuan
Learning robust multi-label hashing for efficient image retrieval
description Supervised hashing generally achieves superior performance over unsupervised or semi-supervised approaches by leveraging semantic labels. However, most existing supervised hashing techniques only deal with image samples with single label. Few of them properly address the practical problem concerning images with multiple labels, which is very common in real applications. In this paper, we seek to address the limitations of the existing schemes by proposing a novel approach, dubbed as Robust Multi-Label Hashing (RMLH). A label hypergraph is constructed to effectively capture high-order semantic correlations of images. And they are preserved into hashing codes with hypergraph consistency and direct label-hashing correlation. Besides, we impose a nuclear norm regularization on correlation matrix to maintain label correlations and robustly accommodate missing labels. Furthermore, an efficient algorithm based on Alternate Direction Method of Multipliers (ADMM) is developed to calculate the optimal hashing codes. Experiments demonstrate that RMLH can outperform state-of-the-art schemes and enjoy much better robustness against missing labels.
format text
author CHEN, Haibao
ZHAO, Yuyan
ZHU, Lei
CHEN, Guilin
SUN, Kaichuan
author_facet CHEN, Haibao
ZHAO, Yuyan
ZHU, Lei
CHEN, Guilin
SUN, Kaichuan
author_sort CHEN, Haibao
title Learning robust multi-label hashing for efficient image retrieval
title_short Learning robust multi-label hashing for efficient image retrieval
title_full Learning robust multi-label hashing for efficient image retrieval
title_fullStr Learning robust multi-label hashing for efficient image retrieval
title_full_unstemmed Learning robust multi-label hashing for efficient image retrieval
title_sort learning robust multi-label hashing for efficient image retrieval
publisher Institutional Knowledge at Singapore Management University
publishDate 2016
url https://ink.library.smu.edu.sg/sis_research/3593
_version_ 1770573339699445760