Ordinal Distance Metric Learning for Image Ranking
Recently, distance metric learning (DML) has attracted much attention in image retrieval, but most previous methods only work for image classification and clustering tasks. In this brief, we focus on designing ordinal DML algorithms for image ranking tasks, by which the rank levels among the images...
Main Authors: | , , , |
---|---|
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
2015
|
Subjects: | |
Online Access: | http://ir.ia.ac.cn/handle/173211/7912 |
id |
ftchiacadsccasia:oai:ir.ia.ac.cn:173211/7912 |
---|---|
record_format |
openpolar |
spelling |
ftchiacadsccasia:oai:ir.ia.ac.cn:173211/7912 2023-07-02T03:32:05+02:00 Ordinal Distance Metric Learning for Image Ranking Li, Changsheng Liu, Qingshan Liu, Jing Lu, Hanqing 2015-07-01 http://ir.ia.ac.cn/handle/173211/7912 英语 eng IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS http://ir.ia.ac.cn/handle/173211/7912 Distance Metric Learning (Dml) Image Ranking Local Geometry Structure Ordinal Relationship Science & Technology Technology AGE ESTIMATION Computer Science Engineering Artificial Intelligence Hardware & Architecture Theory & Methods Electrical & Electronic Article 期刊论文 2015 ftchiacadsccasia 2023-06-13T16:12:11Z Recently, distance metric learning (DML) has attracted much attention in image retrieval, but most previous methods only work for image classification and clustering tasks. In this brief, we focus on designing ordinal DML algorithms for image ranking tasks, by which the rank levels among the images can be well measured. We first present a linear ordinal Mahalanobis DML model that tries to preserve both the local geometry information and the ordinal relationship of the data. Then, we develop a nonlinear DML method by kernelizing the above model, considering of real-world image data with nonlinear structures. To further improve the ranking performance, we finally derive a multiple kernel DML approach inspired by the idea of multiple-kernel learning that performs different kernel operators on different kinds of image features. Extensive experiments on four benchmarks demonstrate the power of the proposed algorithms against some related state-of-the-art methods. Article in Journal/Newspaper DML Institute of Automation: CASIA OpenIR (Chinese Academy of Sciences) |
institution |
Open Polar |
collection |
Institute of Automation: CASIA OpenIR (Chinese Academy of Sciences) |
op_collection_id |
ftchiacadsccasia |
language |
English |
topic |
Distance Metric Learning (Dml) Image Ranking Local Geometry Structure Ordinal Relationship Science & Technology Technology AGE ESTIMATION Computer Science Engineering Artificial Intelligence Hardware & Architecture Theory & Methods Electrical & Electronic |
spellingShingle |
Distance Metric Learning (Dml) Image Ranking Local Geometry Structure Ordinal Relationship Science & Technology Technology AGE ESTIMATION Computer Science Engineering Artificial Intelligence Hardware & Architecture Theory & Methods Electrical & Electronic Li, Changsheng Liu, Qingshan Liu, Jing Lu, Hanqing Ordinal Distance Metric Learning for Image Ranking |
topic_facet |
Distance Metric Learning (Dml) Image Ranking Local Geometry Structure Ordinal Relationship Science & Technology Technology AGE ESTIMATION Computer Science Engineering Artificial Intelligence Hardware & Architecture Theory & Methods Electrical & Electronic |
description |
Recently, distance metric learning (DML) has attracted much attention in image retrieval, but most previous methods only work for image classification and clustering tasks. In this brief, we focus on designing ordinal DML algorithms for image ranking tasks, by which the rank levels among the images can be well measured. We first present a linear ordinal Mahalanobis DML model that tries to preserve both the local geometry information and the ordinal relationship of the data. Then, we develop a nonlinear DML method by kernelizing the above model, considering of real-world image data with nonlinear structures. To further improve the ranking performance, we finally derive a multiple kernel DML approach inspired by the idea of multiple-kernel learning that performs different kernel operators on different kinds of image features. Extensive experiments on four benchmarks demonstrate the power of the proposed algorithms against some related state-of-the-art methods. |
format |
Article in Journal/Newspaper |
author |
Li, Changsheng Liu, Qingshan Liu, Jing Lu, Hanqing |
author_facet |
Li, Changsheng Liu, Qingshan Liu, Jing Lu, Hanqing |
author_sort |
Li, Changsheng |
title |
Ordinal Distance Metric Learning for Image Ranking |
title_short |
Ordinal Distance Metric Learning for Image Ranking |
title_full |
Ordinal Distance Metric Learning for Image Ranking |
title_fullStr |
Ordinal Distance Metric Learning for Image Ranking |
title_full_unstemmed |
Ordinal Distance Metric Learning for Image Ranking |
title_sort |
ordinal distance metric learning for image ranking |
publishDate |
2015 |
url |
http://ir.ia.ac.cn/handle/173211/7912 |
genre |
DML |
genre_facet |
DML |
op_relation |
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS http://ir.ia.ac.cn/handle/173211/7912 |
_version_ |
1770271566790131712 |