Diffraction model-informed neural network for unsupervised layer-based computer-generated holography

Learning-based computer-generated holography (CGH) has shown remarkable promise to enable real-time holographic displays. Supervised CGH requires creating a large-scale dataset with target images and corresponding holograms. We propose a diffraction model-informed neural network framework (self-holo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Optics express 2022-12, Vol.30 (25), p.44814-44826
Hauptverfasser: Shui, Xinghua, Zheng, Huadong, Xia, Xinxing, Yang, Furong, Wang, Weisen, Yu, Yingjie
Format: Artikel
Sprache:eng
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 44826
container_issue 25
container_start_page 44814
container_title Optics express
container_volume 30
creator Shui, Xinghua
Zheng, Huadong
Xia, Xinxing
Yang, Furong
Wang, Weisen
Yu, Yingjie
description Learning-based computer-generated holography (CGH) has shown remarkable promise to enable real-time holographic displays. Supervised CGH requires creating a large-scale dataset with target images and corresponding holograms. We propose a diffraction model-informed neural network framework (self-holo) for 3D phase-only hologram generation. Due to the angular spectrum propagation being incorporated into the neural network, the self-holo can be trained in an unsupervised manner without the need of a labeled dataset. Utilizing the various representations of a 3D object and randomly reconstructing the hologram to one layer of a 3D object keeps the complexity of the self-holo independent of the number of depth layers. The self-holo takes amplitude and depth map images as input and synthesizes a 3D hologram or a 2D hologram. We demonstrate 3D reconstructions with a good 3D effect and the generalizability of self-holo in numerical and optical experiments.
doi_str_mv 10.1364/OE.474137
format Article
fullrecord <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_proquest_miscellaneous_2755576417</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2755576417</sourcerecordid><originalsourceid>FETCH-LOGICAL-c320t-79d6e6a111e0d17eef266be08a6c52c51c6a14d3d3bacc6dc9adee1cf0c411c33</originalsourceid><addsrcrecordid>eNpNkM1OwzAQhC0EoqVw4AVQjnBI8caO3RxRW36kSr3AOXLsTRtI4mAnoL49rloQp52d-TSHIeQa6BSY4Pfr5ZRLDkyekDHQjMeczuTpPz0iF96_UwpcZvKcjJhIk2SWiTExi6osndJ9ZduosQbruGpL6xo0UYuDU3U4_bd1H1Fwo6H1Q4fuq_Ihr9UOXVyovda26YY-vBts0ak-WFtb241T3XZ3Sc5KVXu8Ot4JeXtcvs6f49X66WX-sIo1S2gfy8wIFAoAkBqQiGUiRIF0poROE52CDiE3zLBCaS2MzpRBBF1SzQE0YxNye-jtnP0c0Pd5U3mNda1atIPPE5mmqRQcZEDvDqh21nuHZd65qlFulwPN96Pm62V-GDWwN8faoQi7_JG_K7If18Z0yg</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2755576417</pqid></control><display><type>article</type><title>Diffraction model-informed neural network for unsupervised layer-based computer-generated holography</title><source>DOAJ Directory of Open Access Journals</source><source>Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals</source><source>Alma/SFX Local Collection</source><creator>Shui, Xinghua ; Zheng, Huadong ; Xia, Xinxing ; Yang, Furong ; Wang, Weisen ; Yu, Yingjie</creator><creatorcontrib>Shui, Xinghua ; Zheng, Huadong ; Xia, Xinxing ; Yang, Furong ; Wang, Weisen ; Yu, Yingjie</creatorcontrib><description>Learning-based computer-generated holography (CGH) has shown remarkable promise to enable real-time holographic displays. Supervised CGH requires creating a large-scale dataset with target images and corresponding holograms. We propose a diffraction model-informed neural network framework (self-holo) for 3D phase-only hologram generation. Due to the angular spectrum propagation being incorporated into the neural network, the self-holo can be trained in an unsupervised manner without the need of a labeled dataset. Utilizing the various representations of a 3D object and randomly reconstructing the hologram to one layer of a 3D object keeps the complexity of the self-holo independent of the number of depth layers. The self-holo takes amplitude and depth map images as input and synthesizes a 3D hologram or a 2D hologram. We demonstrate 3D reconstructions with a good 3D effect and the generalizability of self-holo in numerical and optical experiments.</description><identifier>ISSN: 1094-4087</identifier><identifier>EISSN: 1094-4087</identifier><identifier>DOI: 10.1364/OE.474137</identifier><identifier>PMID: 36522896</identifier><language>eng</language><publisher>United States</publisher><ispartof>Optics express, 2022-12, Vol.30 (25), p.44814-44826</ispartof><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c320t-79d6e6a111e0d17eef266be08a6c52c51c6a14d3d3bacc6dc9adee1cf0c411c33</citedby><cites>FETCH-LOGICAL-c320t-79d6e6a111e0d17eef266be08a6c52c51c6a14d3d3bacc6dc9adee1cf0c411c33</cites><orcidid>0000-0002-1361-6782</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,776,780,860,27901,27902</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/36522896$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Shui, Xinghua</creatorcontrib><creatorcontrib>Zheng, Huadong</creatorcontrib><creatorcontrib>Xia, Xinxing</creatorcontrib><creatorcontrib>Yang, Furong</creatorcontrib><creatorcontrib>Wang, Weisen</creatorcontrib><creatorcontrib>Yu, Yingjie</creatorcontrib><title>Diffraction model-informed neural network for unsupervised layer-based computer-generated holography</title><title>Optics express</title><addtitle>Opt Express</addtitle><description>Learning-based computer-generated holography (CGH) has shown remarkable promise to enable real-time holographic displays. Supervised CGH requires creating a large-scale dataset with target images and corresponding holograms. We propose a diffraction model-informed neural network framework (self-holo) for 3D phase-only hologram generation. Due to the angular spectrum propagation being incorporated into the neural network, the self-holo can be trained in an unsupervised manner without the need of a labeled dataset. Utilizing the various representations of a 3D object and randomly reconstructing the hologram to one layer of a 3D object keeps the complexity of the self-holo independent of the number of depth layers. The self-holo takes amplitude and depth map images as input and synthesizes a 3D hologram or a 2D hologram. We demonstrate 3D reconstructions with a good 3D effect and the generalizability of self-holo in numerical and optical experiments.</description><issn>1094-4087</issn><issn>1094-4087</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><recordid>eNpNkM1OwzAQhC0EoqVw4AVQjnBI8caO3RxRW36kSr3AOXLsTRtI4mAnoL49rloQp52d-TSHIeQa6BSY4Pfr5ZRLDkyekDHQjMeczuTpPz0iF96_UwpcZvKcjJhIk2SWiTExi6osndJ9ZduosQbruGpL6xo0UYuDU3U4_bd1H1Fwo6H1Q4fuq_Ihr9UOXVyovda26YY-vBts0ak-WFtb241T3XZ3Sc5KVXu8Ot4JeXtcvs6f49X66WX-sIo1S2gfy8wIFAoAkBqQiGUiRIF0poROE52CDiE3zLBCaS2MzpRBBF1SzQE0YxNye-jtnP0c0Pd5U3mNda1atIPPE5mmqRQcZEDvDqh21nuHZd65qlFulwPN96Pm62V-GDWwN8faoQi7_JG_K7If18Z0yg</recordid><startdate>20221205</startdate><enddate>20221205</enddate><creator>Shui, Xinghua</creator><creator>Zheng, Huadong</creator><creator>Xia, Xinxing</creator><creator>Yang, Furong</creator><creator>Wang, Weisen</creator><creator>Yu, Yingjie</creator><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-1361-6782</orcidid></search><sort><creationdate>20221205</creationdate><title>Diffraction model-informed neural network for unsupervised layer-based computer-generated holography</title><author>Shui, Xinghua ; Zheng, Huadong ; Xia, Xinxing ; Yang, Furong ; Wang, Weisen ; Yu, Yingjie</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c320t-79d6e6a111e0d17eef266be08a6c52c51c6a14d3d3bacc6dc9adee1cf0c411c33</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Shui, Xinghua</creatorcontrib><creatorcontrib>Zheng, Huadong</creatorcontrib><creatorcontrib>Xia, Xinxing</creatorcontrib><creatorcontrib>Yang, Furong</creatorcontrib><creatorcontrib>Wang, Weisen</creatorcontrib><creatorcontrib>Yu, Yingjie</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>MEDLINE - Academic</collection><jtitle>Optics express</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Shui, Xinghua</au><au>Zheng, Huadong</au><au>Xia, Xinxing</au><au>Yang, Furong</au><au>Wang, Weisen</au><au>Yu, Yingjie</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Diffraction model-informed neural network for unsupervised layer-based computer-generated holography</atitle><jtitle>Optics express</jtitle><addtitle>Opt Express</addtitle><date>2022-12-05</date><risdate>2022</risdate><volume>30</volume><issue>25</issue><spage>44814</spage><epage>44826</epage><pages>44814-44826</pages><issn>1094-4087</issn><eissn>1094-4087</eissn><abstract>Learning-based computer-generated holography (CGH) has shown remarkable promise to enable real-time holographic displays. Supervised CGH requires creating a large-scale dataset with target images and corresponding holograms. We propose a diffraction model-informed neural network framework (self-holo) for 3D phase-only hologram generation. Due to the angular spectrum propagation being incorporated into the neural network, the self-holo can be trained in an unsupervised manner without the need of a labeled dataset. Utilizing the various representations of a 3D object and randomly reconstructing the hologram to one layer of a 3D object keeps the complexity of the self-holo independent of the number of depth layers. The self-holo takes amplitude and depth map images as input and synthesizes a 3D hologram or a 2D hologram. We demonstrate 3D reconstructions with a good 3D effect and the generalizability of self-holo in numerical and optical experiments.</abstract><cop>United States</cop><pmid>36522896</pmid><doi>10.1364/OE.474137</doi><tpages>13</tpages><orcidid>https://orcid.org/0000-0002-1361-6782</orcidid><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier ISSN: 1094-4087
ispartof Optics express, 2022-12, Vol.30 (25), p.44814-44826
issn 1094-4087
1094-4087
language eng
recordid cdi_proquest_miscellaneous_2755576417
source DOAJ Directory of Open Access Journals; Elektronische Zeitschriftenbibliothek - Frei zugängliche E-Journals; Alma/SFX Local Collection
title Diffraction model-informed neural network for unsupervised layer-based computer-generated holography
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-21T17%3A11%3A41IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Diffraction%20model-informed%20neural%20network%20for%20unsupervised%20layer-based%20computer-generated%20holography&rft.jtitle=Optics%20express&rft.au=Shui,%20Xinghua&rft.date=2022-12-05&rft.volume=30&rft.issue=25&rft.spage=44814&rft.epage=44826&rft.pages=44814-44826&rft.issn=1094-4087&rft.eissn=1094-4087&rft_id=info:doi/10.1364/OE.474137&rft_dat=%3Cproquest_cross%3E2755576417%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2755576417&rft_id=info:pmid/36522896&rfr_iscdi=true