Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval
A rich line of works focus on designing elegant loss functions under the deep metric learning (DML) paradigm to learn a discriminative embedding space for remote sensing image retrieval (RSIR). Essentially, such embedding space could efficiently distinguish deep feature descriptors. So far, most exi...
Published in: | Sensors |
---|---|
Main Authors: | , , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2020
|
Subjects: | |
Online Access: | https://doi.org/10.3390/s20010291 |
id |
ftmdpi:oai:mdpi.com:/1424-8220/20/1/291/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/1424-8220/20/1/291/ 2023-08-20T04:06:09+02:00 Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval Pingping Liu Guixia Gou Xue Shan Dan Tao Qiuzhan Zhou 2020-01-04 application/pdf https://doi.org/10.3390/s20010291 EN eng Multidisciplinary Digital Publishing Institute Intelligent Sensors https://dx.doi.org/10.3390/s20010291 https://creativecommons.org/licenses/by/4.0/ Sensors; Volume 20; Issue 1; Pages: 291 remote sensing image retrieval convolutional neural network deep metric learning global optimization Text 2020 ftmdpi https://doi.org/10.3390/s20010291 2023-07-31T22:58:02Z A rich line of works focus on designing elegant loss functions under the deep metric learning (DML) paradigm to learn a discriminative embedding space for remote sensing image retrieval (RSIR). Essentially, such embedding space could efficiently distinguish deep feature descriptors. So far, most existing losses used in RSIR are based on triplets, which have disadvantages of local optimization, slow convergence and insufficient use of similarity structure in a mini-batch. In this paper, we present a novel DML method named as global optimal structured loss to deal with the limitation of triplet loss. To be specific, we use a softmax function rather than a hinge function in our novel loss to realize global optimization. In addition, we present a novel optimal structured loss, which globally learn an efficient deep embedding space with mined informative sample pairs to force the positive pairs within a limitation and push the negative ones far away from a given boundary. We have conducted extensive experiments on four public remote sensing datasets and the results show that the proposed global optimal structured loss with pairs mining scheme achieves the state-of-the-art performance compared with the baselines. Text DML MDPI Open Access Publishing Triplets ENVELOPE(-59.750,-59.750,-62.383,-62.383) Sensors 20 1 291 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
remote sensing image retrieval convolutional neural network deep metric learning global optimization |
spellingShingle |
remote sensing image retrieval convolutional neural network deep metric learning global optimization Pingping Liu Guixia Gou Xue Shan Dan Tao Qiuzhan Zhou Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval |
topic_facet |
remote sensing image retrieval convolutional neural network deep metric learning global optimization |
description |
A rich line of works focus on designing elegant loss functions under the deep metric learning (DML) paradigm to learn a discriminative embedding space for remote sensing image retrieval (RSIR). Essentially, such embedding space could efficiently distinguish deep feature descriptors. So far, most existing losses used in RSIR are based on triplets, which have disadvantages of local optimization, slow convergence and insufficient use of similarity structure in a mini-batch. In this paper, we present a novel DML method named as global optimal structured loss to deal with the limitation of triplet loss. To be specific, we use a softmax function rather than a hinge function in our novel loss to realize global optimization. In addition, we present a novel optimal structured loss, which globally learn an efficient deep embedding space with mined informative sample pairs to force the positive pairs within a limitation and push the negative ones far away from a given boundary. We have conducted extensive experiments on four public remote sensing datasets and the results show that the proposed global optimal structured loss with pairs mining scheme achieves the state-of-the-art performance compared with the baselines. |
format |
Text |
author |
Pingping Liu Guixia Gou Xue Shan Dan Tao Qiuzhan Zhou |
author_facet |
Pingping Liu Guixia Gou Xue Shan Dan Tao Qiuzhan Zhou |
author_sort |
Pingping Liu |
title |
Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval |
title_short |
Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval |
title_full |
Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval |
title_fullStr |
Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval |
title_full_unstemmed |
Global Optimal Structured Embedding Learning for Remote Sensing Image Retrieval |
title_sort |
global optimal structured embedding learning for remote sensing image retrieval |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2020 |
url |
https://doi.org/10.3390/s20010291 |
long_lat |
ENVELOPE(-59.750,-59.750,-62.383,-62.383) |
geographic |
Triplets |
geographic_facet |
Triplets |
genre |
DML |
genre_facet |
DML |
op_source |
Sensors; Volume 20; Issue 1; Pages: 291 |
op_relation |
Intelligent Sensors https://dx.doi.org/10.3390/s20010291 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/s20010291 |
container_title |
Sensors |
container_volume |
20 |
container_issue |
1 |
container_start_page |
291 |
_version_ |
1774717086705123328 |