Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation
In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of eac...
Gespeichert in:
Veröffentlicht in: | arXiv.org 2021-05 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | |
---|---|
container_issue | |
container_start_page | |
container_title | arXiv.org |
container_volume | |
creator | Kordopatis-Zilos, Giorgos Galopoulos, Panagiotis Papadopoulos, Symeon Kompatsiaris, Ioannis |
description | In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of each approach with two different modules. The first leverages the EfficientNet architecture to assign images to a specific geographic cell in a robust way. The second introduces a new residual architecture that is trained with contrastive learning to map input images to an embedding space that minimizes the pairwise geodesic distance of same-location images. For the final location estimation, the two modules are combined with a search-within-cell scheme, where the locations of most similar images from the predicted geographic cell are aggregated based on a spatial clustering scheme. Our approach demonstrates very competitive performance on four public datasets, achieving new state-of-the-art performance in fine granularity scales, i.e., 15.0% at 1km range on Im2GPS3k. |
format | Article |
fullrecord | <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2528647553</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2528647553</sourcerecordid><originalsourceid>FETCH-proquest_journals_25286475533</originalsourceid><addsrcrecordid>eNqNjssKwjAURIMgWLT_EHBdqOlzK6XqorhybbnGm5ISE03Sfr9R_ABXc2DOwCxIxLJsl9Q5YysSOzemacrKihVFFpFrhzNaGKQeaCuE5BK1P6OnoO-0MdpbcF7OSDsEqz-WMJbuOZ8seKRHZW6gEsdBBcVw8NJo2obJ44sbshSgHMa_XJPtob00p-RpzWtC5_vRTFaHqmcFq8u8Cq-y_6w3awVEWQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2528647553</pqid></control><display><type>article</type><title>Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation</title><source>Free E- Journals</source><creator>Kordopatis-Zilos, Giorgos ; Galopoulos, Panagiotis ; Papadopoulos, Symeon ; Kompatsiaris, Ioannis</creator><creatorcontrib>Kordopatis-Zilos, Giorgos ; Galopoulos, Panagiotis ; Papadopoulos, Symeon ; Kompatsiaris, Ioannis</creatorcontrib><description>In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of each approach with two different modules. The first leverages the EfficientNet architecture to assign images to a specific geographic cell in a robust way. The second introduces a new residual architecture that is trained with contrastive learning to map input images to an embedding space that minimizes the pairwise geodesic distance of same-location images. For the final location estimation, the two modules are combined with a search-within-cell scheme, where the locations of most similar images from the predicted geographic cell are aggregated based on a spatial clustering scheme. Our approach demonstrates very competitive performance on four public datasets, achieving new state-of-the-art performance in fine granularity scales, i.e., 15.0% at 1km range on Im2GPS3k.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Clustering ; Image classification ; Learning ; Modules ; Retrieval</subject><ispartof>arXiv.org, 2021-05</ispartof><rights>2021. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>776,780</link.rule.ids></links><search><creatorcontrib>Kordopatis-Zilos, Giorgos</creatorcontrib><creatorcontrib>Galopoulos, Panagiotis</creatorcontrib><creatorcontrib>Papadopoulos, Symeon</creatorcontrib><creatorcontrib>Kompatsiaris, Ioannis</creatorcontrib><title>Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation</title><title>arXiv.org</title><description>In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of each approach with two different modules. The first leverages the EfficientNet architecture to assign images to a specific geographic cell in a robust way. The second introduces a new residual architecture that is trained with contrastive learning to map input images to an embedding space that minimizes the pairwise geodesic distance of same-location images. For the final location estimation, the two modules are combined with a search-within-cell scheme, where the locations of most similar images from the predicted geographic cell are aggregated based on a spatial clustering scheme. Our approach demonstrates very competitive performance on four public datasets, achieving new state-of-the-art performance in fine granularity scales, i.e., 15.0% at 1km range on Im2GPS3k.</description><subject>Clustering</subject><subject>Image classification</subject><subject>Learning</subject><subject>Modules</subject><subject>Retrieval</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2021</creationdate><recordtype>article</recordtype><sourceid>BENPR</sourceid><recordid>eNqNjssKwjAURIMgWLT_EHBdqOlzK6XqorhybbnGm5ISE03Sfr9R_ABXc2DOwCxIxLJsl9Q5YysSOzemacrKihVFFpFrhzNaGKQeaCuE5BK1P6OnoO-0MdpbcF7OSDsEqz-WMJbuOZ8seKRHZW6gEsdBBcVw8NJo2obJ44sbshSgHMa_XJPtob00p-RpzWtC5_vRTFaHqmcFq8u8Cq-y_6w3awVEWQ</recordid><startdate>20210517</startdate><enddate>20210517</enddate><creator>Kordopatis-Zilos, Giorgos</creator><creator>Galopoulos, Panagiotis</creator><creator>Papadopoulos, Symeon</creator><creator>Kompatsiaris, Ioannis</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20210517</creationdate><title>Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation</title><author>Kordopatis-Zilos, Giorgos ; Galopoulos, Panagiotis ; Papadopoulos, Symeon ; Kompatsiaris, Ioannis</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_25286475533</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2021</creationdate><topic>Clustering</topic><topic>Image classification</topic><topic>Learning</topic><topic>Modules</topic><topic>Retrieval</topic><toplevel>online_resources</toplevel><creatorcontrib>Kordopatis-Zilos, Giorgos</creatorcontrib><creatorcontrib>Galopoulos, Panagiotis</creatorcontrib><creatorcontrib>Papadopoulos, Symeon</creatorcontrib><creatorcontrib>Kompatsiaris, Ioannis</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science & Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Kordopatis-Zilos, Giorgos</au><au>Galopoulos, Panagiotis</au><au>Papadopoulos, Symeon</au><au>Kompatsiaris, Ioannis</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation</atitle><jtitle>arXiv.org</jtitle><date>2021-05-17</date><risdate>2021</risdate><eissn>2331-8422</eissn><abstract>In this paper, we address the problem of global-scale image geolocation, proposing a mixed classification-retrieval scheme. Unlike other methods that strictly tackle the problem as a classification or retrieval task, we combine the two practices in a unified solution leveraging the advantages of each approach with two different modules. The first leverages the EfficientNet architecture to assign images to a specific geographic cell in a robust way. The second introduces a new residual architecture that is trained with contrastive learning to map input images to an embedding space that minimizes the pairwise geodesic distance of same-location images. For the final location estimation, the two modules are combined with a search-within-cell scheme, where the locations of most similar images from the predicted geographic cell are aggregated based on a spatial clustering scheme. Our approach demonstrates very competitive performance on four public datasets, achieving new state-of-the-art performance in fine granularity scales, i.e., 15.0% at 1km range on Im2GPS3k.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | EISSN: 2331-8422 |
ispartof | arXiv.org, 2021-05 |
issn | 2331-8422 |
language | eng |
recordid | cdi_proquest_journals_2528647553 |
source | Free E- Journals |
subjects | Clustering Image classification Learning Modules Retrieval |
title | Leveraging EfficientNet and Contrastive Learning for Accurate Global-scale Location Estimation |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-11T08%3A28%3A00IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Leveraging%20EfficientNet%20and%20Contrastive%20Learning%20for%20Accurate%20Global-scale%20Location%20Estimation&rft.jtitle=arXiv.org&rft.au=Kordopatis-Zilos,%20Giorgos&rft.date=2021-05-17&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2528647553%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2528647553&rft_id=info:pmid/&rfr_iscdi=true |