Discrete Spherical Image Representation for CNN-Based Inclination Estimation

How an image is represented as the input of a convolutional neural network (CNN) is important because this input directly influences the performance of the CNN. In this paper, we investigate the representation of spherical images by focusing on the inclination estimation of a spherical camera. Unlik...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE access 2020, Vol.8, p.2008-2022
Hauptverfasser: Shan, Yuhao, Li, Shigang
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 2022
container_issue
container_start_page 2008
container_title IEEE access
container_volume 8
creator Shan, Yuhao
Li, Shigang
description How an image is represented as the input of a convolutional neural network (CNN) is important because this input directly influences the performance of the CNN. In this paper, we investigate the representation of spherical images by focusing on the inclination estimation of a spherical camera. Unlike other approaches to CNN-based inclination estimation, a spherical image is represented as a geodesic-division-based discrete spherical image (DSI) that is obtained by sampling a sphere as uniformly as possible. The input of the CNN is a single image that consists of five parallelograms flattened from a regular icosahedron. To demonstrate the advantage of the proposed method, comparative experiments are conducted with two other spherical image representations, namely, equirectangular projection (ERP) and cubemap projection (CMP). The experimental results show that the proposed method using a geodesic-division-based discrete spherical image as the CNN input obtains the best performance-better than that of the cubemap and far superior to that of the equirectangular image. The effect of the image representations used becomes more significant as the relative inclination decreases. Moreover, comparative experiments are conducted using the state-of-the-art methods for spherical camera inclination compensation to further illustrate the superiority of the DSI representation. Consequently, the proposed method provides an important reference for the development of CNNs intended for spherical images.
doi_str_mv 10.1109/ACCESS.2019.2962133
format Article
fullrecord <record><control><sourceid>proquest_ieee_</sourceid><recordid>TN_cdi_proquest_journals_2454717037</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>8943206</ieee_id><doaj_id>oai_doaj_org_article_5bc017b582b646ae98b9e4c97845f4c4</doaj_id><sourcerecordid>2454717037</sourcerecordid><originalsourceid>FETCH-LOGICAL-c474t-4f36fb8f9868c46419b267b14fda389672625acfc55fb15ac04ac116471bcc633</originalsourceid><addsrcrecordid>eNpNUctKA0EQXETBoH6BlwXPG-f9OMY1aiBEMHoeZiY9uiHZjTObg3_vmA3BvnTR3VXVUEVxi9EYY6TvJ3U9XS7HBGE9JloQTOlZMSJY6IpyKs7_4cviJqU1yqXyiMtRMX9sko_QQ7ncfUFsvN2Us639hPINdhEStL3tm64tQxfLerGoHmyCVTlr_aZph8009c32AK-Li2A3CW6O_ar4eJq-1y_V_PV5Vk_mlWeS9RULVASnglZCeSYY1o4I6TALK0uVFpIIwq0PnvPgcEaIWY-xYBI77wWlV8Vs0F11dm12MdvHH9PZxhwGXfw0NvaN34DhziMsHVfECSYsaOU0MK-lYjwwz7LW3aC1i933HlJv1t0-tvl9QxjPlhJRma_ocOVjl1KEcHLFyPylYIYUzF8K5phCZt0OrAYATgylGSVI0F9vvoHg</addsrcrecordid><sourcetype>Open Website</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2454717037</pqid></control><display><type>article</type><title>Discrete Spherical Image Representation for CNN-Based Inclination Estimation</title><source>IEEE Open Access Journals</source><source>DOAJ Directory of Open Access Journals</source><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><creator>Shan, Yuhao ; Li, Shigang</creator><creatorcontrib>Shan, Yuhao ; Li, Shigang</creatorcontrib><description>How an image is represented as the input of a convolutional neural network (CNN) is important because this input directly influences the performance of the CNN. In this paper, we investigate the representation of spherical images by focusing on the inclination estimation of a spherical camera. Unlike other approaches to CNN-based inclination estimation, a spherical image is represented as a geodesic-division-based discrete spherical image (DSI) that is obtained by sampling a sphere as uniformly as possible. The input of the CNN is a single image that consists of five parallelograms flattened from a regular icosahedron. To demonstrate the advantage of the proposed method, comparative experiments are conducted with two other spherical image representations, namely, equirectangular projection (ERP) and cubemap projection (CMP). The experimental results show that the proposed method using a geodesic-division-based discrete spherical image as the CNN input obtains the best performance-better than that of the cubemap and far superior to that of the equirectangular image. The effect of the image representations used becomes more significant as the relative inclination decreases. Moreover, comparative experiments are conducted using the state-of-the-art methods for spherical camera inclination compensation to further illustrate the superiority of the DSI representation. Consequently, the proposed method provides an important reference for the development of CNNs intended for spherical images.</description><identifier>ISSN: 2169-3536</identifier><identifier>EISSN: 2169-3536</identifier><identifier>DOI: 10.1109/ACCESS.2019.2962133</identifier><identifier>CODEN: IAECCG</identifier><language>eng</language><publisher>Piscataway: IEEE</publisher><subject>Artificial neural networks ; Cameras ; Computer vision ; Convolutional neural networks ; Estimation ; Icosahedrons ; Image representation ; Inclination ; inclination estimation ; Parallelograms ; Representations ; Solid modeling ; spherical images ; spherical representation ; Task analysis</subject><ispartof>IEEE access, 2020, Vol.8, p.2008-2022</ispartof><rights>Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2020</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c474t-4f36fb8f9868c46419b267b14fda389672625acfc55fb15ac04ac116471bcc633</citedby><cites>FETCH-LOGICAL-c474t-4f36fb8f9868c46419b267b14fda389672625acfc55fb15ac04ac116471bcc633</cites><orcidid>0000-0003-4821-326X ; 0000-0001-7197-5495</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/8943206$$EHTML$$P50$$Gieee$$Hfree_for_read</linktohtml><link.rule.ids>314,776,780,860,2095,4009,27612,27902,27903,27904,54911</link.rule.ids></links><search><creatorcontrib>Shan, Yuhao</creatorcontrib><creatorcontrib>Li, Shigang</creatorcontrib><title>Discrete Spherical Image Representation for CNN-Based Inclination Estimation</title><title>IEEE access</title><addtitle>Access</addtitle><description>How an image is represented as the input of a convolutional neural network (CNN) is important because this input directly influences the performance of the CNN. In this paper, we investigate the representation of spherical images by focusing on the inclination estimation of a spherical camera. Unlike other approaches to CNN-based inclination estimation, a spherical image is represented as a geodesic-division-based discrete spherical image (DSI) that is obtained by sampling a sphere as uniformly as possible. The input of the CNN is a single image that consists of five parallelograms flattened from a regular icosahedron. To demonstrate the advantage of the proposed method, comparative experiments are conducted with two other spherical image representations, namely, equirectangular projection (ERP) and cubemap projection (CMP). The experimental results show that the proposed method using a geodesic-division-based discrete spherical image as the CNN input obtains the best performance-better than that of the cubemap and far superior to that of the equirectangular image. The effect of the image representations used becomes more significant as the relative inclination decreases. Moreover, comparative experiments are conducted using the state-of-the-art methods for spherical camera inclination compensation to further illustrate the superiority of the DSI representation. Consequently, the proposed method provides an important reference for the development of CNNs intended for spherical images.</description><subject>Artificial neural networks</subject><subject>Cameras</subject><subject>Computer vision</subject><subject>Convolutional neural networks</subject><subject>Estimation</subject><subject>Icosahedrons</subject><subject>Image representation</subject><subject>Inclination</subject><subject>inclination estimation</subject><subject>Parallelograms</subject><subject>Representations</subject><subject>Solid modeling</subject><subject>spherical images</subject><subject>spherical representation</subject><subject>Task analysis</subject><issn>2169-3536</issn><issn>2169-3536</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2020</creationdate><recordtype>article</recordtype><sourceid>ESBDL</sourceid><sourceid>RIE</sourceid><sourceid>DOA</sourceid><recordid>eNpNUctKA0EQXETBoH6BlwXPG-f9OMY1aiBEMHoeZiY9uiHZjTObg3_vmA3BvnTR3VXVUEVxi9EYY6TvJ3U9XS7HBGE9JloQTOlZMSJY6IpyKs7_4cviJqU1yqXyiMtRMX9sko_QQ7ncfUFsvN2Us639hPINdhEStL3tm64tQxfLerGoHmyCVTlr_aZph8009c32AK-Li2A3CW6O_ar4eJq-1y_V_PV5Vk_mlWeS9RULVASnglZCeSYY1o4I6TALK0uVFpIIwq0PnvPgcEaIWY-xYBI77wWlV8Vs0F11dm12MdvHH9PZxhwGXfw0NvaN34DhziMsHVfECSYsaOU0MK-lYjwwz7LW3aC1i933HlJv1t0-tvl9QxjPlhJRma_ocOVjl1KEcHLFyPylYIYUzF8K5phCZt0OrAYATgylGSVI0F9vvoHg</recordid><startdate>2020</startdate><enddate>2020</enddate><creator>Shan, Yuhao</creator><creator>Li, Shigang</creator><general>IEEE</general><general>The Institute of Electrical and Electronics Engineers, Inc. (IEEE)</general><scope>97E</scope><scope>ESBDL</scope><scope>RIA</scope><scope>RIE</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SC</scope><scope>7SP</scope><scope>7SR</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope><scope>DOA</scope><orcidid>https://orcid.org/0000-0003-4821-326X</orcidid><orcidid>https://orcid.org/0000-0001-7197-5495</orcidid></search><sort><creationdate>2020</creationdate><title>Discrete Spherical Image Representation for CNN-Based Inclination Estimation</title><author>Shan, Yuhao ; Li, Shigang</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c474t-4f36fb8f9868c46419b267b14fda389672625acfc55fb15ac04ac116471bcc633</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2020</creationdate><topic>Artificial neural networks</topic><topic>Cameras</topic><topic>Computer vision</topic><topic>Convolutional neural networks</topic><topic>Estimation</topic><topic>Icosahedrons</topic><topic>Image representation</topic><topic>Inclination</topic><topic>inclination estimation</topic><topic>Parallelograms</topic><topic>Representations</topic><topic>Solid modeling</topic><topic>spherical images</topic><topic>spherical representation</topic><topic>Task analysis</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shan, Yuhao</creatorcontrib><creatorcontrib>Li, Shigang</creatorcontrib><collection>IEEE All-Society Periodicals Package (ASPP) 2005-present</collection><collection>IEEE Open Access Journals</collection><collection>IEEE All-Society Periodicals Package (ASPP) 1998-Present</collection><collection>IEEE Electronic Library (IEL)</collection><collection>CrossRef</collection><collection>Computer and Information Systems Abstracts</collection><collection>Electronics &amp; Communications Abstracts</collection><collection>Engineered Materials Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection><collection>DOAJ Directory of Open Access Journals</collection><jtitle>IEEE access</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shan, Yuhao</au><au>Li, Shigang</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Discrete Spherical Image Representation for CNN-Based Inclination Estimation</atitle><jtitle>IEEE access</jtitle><stitle>Access</stitle><date>2020</date><risdate>2020</risdate><volume>8</volume><spage>2008</spage><epage>2022</epage><pages>2008-2022</pages><issn>2169-3536</issn><eissn>2169-3536</eissn><coden>IAECCG</coden><abstract>How an image is represented as the input of a convolutional neural network (CNN) is important because this input directly influences the performance of the CNN. In this paper, we investigate the representation of spherical images by focusing on the inclination estimation of a spherical camera. Unlike other approaches to CNN-based inclination estimation, a spherical image is represented as a geodesic-division-based discrete spherical image (DSI) that is obtained by sampling a sphere as uniformly as possible. The input of the CNN is a single image that consists of five parallelograms flattened from a regular icosahedron. To demonstrate the advantage of the proposed method, comparative experiments are conducted with two other spherical image representations, namely, equirectangular projection (ERP) and cubemap projection (CMP). The experimental results show that the proposed method using a geodesic-division-based discrete spherical image as the CNN input obtains the best performance-better than that of the cubemap and far superior to that of the equirectangular image. The effect of the image representations used becomes more significant as the relative inclination decreases. Moreover, comparative experiments are conducted using the state-of-the-art methods for spherical camera inclination compensation to further illustrate the superiority of the DSI representation. Consequently, the proposed method provides an important reference for the development of CNNs intended for spherical images.</abstract><cop>Piscataway</cop><pub>IEEE</pub><doi>10.1109/ACCESS.2019.2962133</doi><tpages>15</tpages><orcidid>https://orcid.org/0000-0003-4821-326X</orcidid><orcidid>https://orcid.org/0000-0001-7197-5495</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 2169-3536
ispartof IEEE access, 2020, Vol.8, p.2008-2022
issn 2169-3536
2169-3536
language eng
recordid cdi_proquest_journals_2454717037
source IEEE Open Access Journals; DOAJ Directory of Open Access Journals; Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals
subjects Artificial neural networks
Cameras
Computer vision
Convolutional neural networks
Estimation
Icosahedrons
Image representation
Inclination
inclination estimation
Parallelograms
Representations
Solid modeling
spherical images
spherical representation
Task analysis
title Discrete Spherical Image Representation for CNN-Based Inclination Estimation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T08%3A00%3A52IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_ieee_&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Discrete%20Spherical%20Image%20Representation%20for%20CNN-Based%20Inclination%20Estimation&rft.jtitle=IEEE%20access&rft.au=Shan,%20Yuhao&rft.date=2020&rft.volume=8&rft.spage=2008&rft.epage=2022&rft.pages=2008-2022&rft.issn=2169-3536&rft.eissn=2169-3536&rft.coden=IAECCG&rft_id=info:doi/10.1109/ACCESS.2019.2962133&rft_dat=%3Cproquest_ieee_%3E2454717037%3C/proquest_ieee_%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2454717037&rft_id=info:pmid/&rft_ieee_id=8943206&rft_doaj_id=oai_doaj_org_article_5bc017b582b646ae98b9e4c97845f4c4&rfr_iscdi=true