Revisiting Event-based Video Frame Interpolation

Dynamic vision sensors or event cameras provide rich complementary information for video frame interpolation. Existing state-of-the-art methods follow the paradigm of combining both synthesis-based and warping networks. However, few of those methods fully respect the intrinsic characteristics of eve...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-07
Hauptverfasser: Chen, Jiaben, Zhu, Yichen, Lian, Dongze, Yang, Jiaqi, Wang, Yifu, Zhang, Renrui, Liu, Xinhang, Qian, Shenhan, Kneip, Laurent, Gao, Shenghua
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 Chen, Jiaben
Zhu, Yichen
Lian, Dongze
Yang, Jiaqi
Wang, Yifu
Zhang, Renrui
Liu, Xinhang
Qian, Shenhan
Kneip, Laurent
Gao, Shenghua
description Dynamic vision sensors or event cameras provide rich complementary information for video frame interpolation. Existing state-of-the-art methods follow the paradigm of combining both synthesis-based and warping networks. However, few of those methods fully respect the intrinsic characteristics of events streams. Given that event cameras only encode intensity changes and polarity rather than color intensities, estimating optical flow from events is arguably more difficult than from RGB information. We therefore propose to incorporate RGB information in an event-guided optical flow refinement strategy. Moreover, in light of the quasi-continuous nature of the time signals provided by event cameras, we propose a divide-and-conquer strategy in which event-based intermediate frame synthesis happens incrementally in multiple simplified stages rather than in a single, long stage. Extensive experiments on both synthetic and real-world datasets show that these modifications lead to more reliable and realistic intermediate frame results than previous video frame interpolation methods. Our findings underline that a careful consideration of event characteristics such as high temporal density and elevated noise benefits interpolation accuracy.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2841686320</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2841686320</sourcerecordid><originalsourceid>FETCH-proquest_journals_28416863203</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwCEotyyzOLMnMS1dwLUvNK9FNSixOTVEIy0xJzVdwK0rMTVXwzCtJLSrIz0ksyczP42FgTUvMKU7lhdLcDMpuriHOHroFRfmFpanFJfFZ-aVFeUCpeCMLE0MzCzNjIwNj4lQBAGzfMyE</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2841686320</pqid></control><display><type>article</type><title>Revisiting Event-based Video Frame Interpolation</title><source>Freely Accessible Journals</source><creator>Chen, Jiaben ; Zhu, Yichen ; Lian, Dongze ; Yang, Jiaqi ; Wang, Yifu ; Zhang, Renrui ; Liu, Xinhang ; Qian, Shenhan ; Kneip, Laurent ; Gao, Shenghua</creator><creatorcontrib>Chen, Jiaben ; Zhu, Yichen ; Lian, Dongze ; Yang, Jiaqi ; Wang, Yifu ; Zhang, Renrui ; Liu, Xinhang ; Qian, Shenhan ; Kneip, Laurent ; Gao, Shenghua</creatorcontrib><description>Dynamic vision sensors or event cameras provide rich complementary information for video frame interpolation. Existing state-of-the-art methods follow the paradigm of combining both synthesis-based and warping networks. However, few of those methods fully respect the intrinsic characteristics of events streams. Given that event cameras only encode intensity changes and polarity rather than color intensities, estimating optical flow from events is arguably more difficult than from RGB information. We therefore propose to incorporate RGB information in an event-guided optical flow refinement strategy. Moreover, in light of the quasi-continuous nature of the time signals provided by event cameras, we propose a divide-and-conquer strategy in which event-based intermediate frame synthesis happens incrementally in multiple simplified stages rather than in a single, long stage. Extensive experiments on both synthetic and real-world datasets show that these modifications lead to more reliable and realistic intermediate frame results than previous video frame interpolation methods. Our findings underline that a careful consideration of event characteristics such as high temporal density and elevated noise benefits interpolation accuracy.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Cameras ; Interpolation ; Optical flow (image analysis) ; Synthesis ; Time signals</subject><ispartof>arXiv.org, 2023-07</ispartof><rights>2023. 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>780,784</link.rule.ids></links><search><creatorcontrib>Chen, Jiaben</creatorcontrib><creatorcontrib>Zhu, Yichen</creatorcontrib><creatorcontrib>Lian, Dongze</creatorcontrib><creatorcontrib>Yang, Jiaqi</creatorcontrib><creatorcontrib>Wang, Yifu</creatorcontrib><creatorcontrib>Zhang, Renrui</creatorcontrib><creatorcontrib>Liu, Xinhang</creatorcontrib><creatorcontrib>Qian, Shenhan</creatorcontrib><creatorcontrib>Kneip, Laurent</creatorcontrib><creatorcontrib>Gao, Shenghua</creatorcontrib><title>Revisiting Event-based Video Frame Interpolation</title><title>arXiv.org</title><description>Dynamic vision sensors or event cameras provide rich complementary information for video frame interpolation. Existing state-of-the-art methods follow the paradigm of combining both synthesis-based and warping networks. However, few of those methods fully respect the intrinsic characteristics of events streams. Given that event cameras only encode intensity changes and polarity rather than color intensities, estimating optical flow from events is arguably more difficult than from RGB information. We therefore propose to incorporate RGB information in an event-guided optical flow refinement strategy. Moreover, in light of the quasi-continuous nature of the time signals provided by event cameras, we propose a divide-and-conquer strategy in which event-based intermediate frame synthesis happens incrementally in multiple simplified stages rather than in a single, long stage. Extensive experiments on both synthetic and real-world datasets show that these modifications lead to more reliable and realistic intermediate frame results than previous video frame interpolation methods. Our findings underline that a careful consideration of event characteristics such as high temporal density and elevated noise benefits interpolation accuracy.</description><subject>Cameras</subject><subject>Interpolation</subject><subject>Optical flow (image analysis)</subject><subject>Synthesis</subject><subject>Time signals</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwCEotyyzOLMnMS1dwLUvNK9FNSixOTVEIy0xJzVdwK0rMTVXwzCtJLSrIz0ksyczP42FgTUvMKU7lhdLcDMpuriHOHroFRfmFpanFJfFZ-aVFeUCpeCMLE0MzCzNjIwNj4lQBAGzfMyE</recordid><startdate>20230724</startdate><enddate>20230724</enddate><creator>Chen, Jiaben</creator><creator>Zhu, Yichen</creator><creator>Lian, Dongze</creator><creator>Yang, Jiaqi</creator><creator>Wang, Yifu</creator><creator>Zhang, Renrui</creator><creator>Liu, Xinhang</creator><creator>Qian, Shenhan</creator><creator>Kneip, Laurent</creator><creator>Gao, Shenghua</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>20230724</creationdate><title>Revisiting Event-based Video Frame Interpolation</title><author>Chen, Jiaben ; Zhu, Yichen ; Lian, Dongze ; Yang, Jiaqi ; Wang, Yifu ; Zhang, Renrui ; Liu, Xinhang ; Qian, Shenhan ; Kneip, Laurent ; Gao, Shenghua</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28416863203</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Cameras</topic><topic>Interpolation</topic><topic>Optical flow (image analysis)</topic><topic>Synthesis</topic><topic>Time signals</topic><toplevel>online_resources</toplevel><creatorcontrib>Chen, Jiaben</creatorcontrib><creatorcontrib>Zhu, Yichen</creatorcontrib><creatorcontrib>Lian, Dongze</creatorcontrib><creatorcontrib>Yang, Jiaqi</creatorcontrib><creatorcontrib>Wang, Yifu</creatorcontrib><creatorcontrib>Zhang, Renrui</creatorcontrib><creatorcontrib>Liu, Xinhang</creatorcontrib><creatorcontrib>Qian, Shenhan</creatorcontrib><creatorcontrib>Kneip, Laurent</creatorcontrib><creatorcontrib>Gao, Shenghua</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; 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>Chen, Jiaben</au><au>Zhu, Yichen</au><au>Lian, Dongze</au><au>Yang, Jiaqi</au><au>Wang, Yifu</au><au>Zhang, Renrui</au><au>Liu, Xinhang</au><au>Qian, Shenhan</au><au>Kneip, Laurent</au><au>Gao, Shenghua</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Revisiting Event-based Video Frame Interpolation</atitle><jtitle>arXiv.org</jtitle><date>2023-07-24</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Dynamic vision sensors or event cameras provide rich complementary information for video frame interpolation. Existing state-of-the-art methods follow the paradigm of combining both synthesis-based and warping networks. However, few of those methods fully respect the intrinsic characteristics of events streams. Given that event cameras only encode intensity changes and polarity rather than color intensities, estimating optical flow from events is arguably more difficult than from RGB information. We therefore propose to incorporate RGB information in an event-guided optical flow refinement strategy. Moreover, in light of the quasi-continuous nature of the time signals provided by event cameras, we propose a divide-and-conquer strategy in which event-based intermediate frame synthesis happens incrementally in multiple simplified stages rather than in a single, long stage. Extensive experiments on both synthetic and real-world datasets show that these modifications lead to more reliable and realistic intermediate frame results than previous video frame interpolation methods. Our findings underline that a careful consideration of event characteristics such as high temporal density and elevated noise benefits interpolation accuracy.</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, 2023-07
issn 2331-8422
language eng
recordid cdi_proquest_journals_2841686320
source Freely Accessible Journals
subjects Cameras
Interpolation
Optical flow (image analysis)
Synthesis
Time signals
title Revisiting Event-based Video Frame Interpolation
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-28T18%3A20%3A10IST&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=Revisiting%20Event-based%20Video%20Frame%20Interpolation&rft.jtitle=arXiv.org&rft.au=Chen,%20Jiaben&rft.date=2023-07-24&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2841686320%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2841686320&rft_id=info:pmid/&rfr_iscdi=true