Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]

Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]

ID:139794

大小:14.07 MB

页数:16页

时间:2017-06-26

Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]_第1页
Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]_第2页
Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]_第3页
Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]_第4页
Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]_第5页
资源描述:

《Large-scale Crowdsourced Study for Tone Mapped HDR Pictures.[2017][IEEE Trans Image Process][10.1109tip.2017.2713945]》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、Thisarticlehasbeenacceptedforpublicationinafutureissueofthisjournal,buthasnotbeenfullyedited.Contentmaychangepriortofinalpublication.Citationinformation:DOI10.1109/TIP.2017.2713945,IEEETransactionsonImageProcessing1Large-scaleCrowdsourcedStudyforToneMappedHDRPicturesDebarat

2、iKundu,DeeptiGhadiyaram,StudentMember,IEEE,AlanC.BovikFellow,IEEE,andBrianL.EvansFellow,IEEEAbstract—Measuringdigitalpicturequality,asperceivedbysmartphonesanddigitalSLRcamerascancreateaestheticallyhumanobservers,isincreasinglyimportantinmanyapplica-pleasingimagesbyfusingam

3、ultiply-exposedstackofimages.tionsinwhichhumansaretheultimateconsumersofvisualHDRimages,commonlyrepresentedby16or32information.Standarddynamicrange(SDR)imagesprovidebits/color/pixel,typicallyareobtainedbyblendingastackof8bits/color/pixel.Highdynamicrange(HDR)images,usuallyc

4、reatedfrommultipleexposuresofthesamescene,canprovideSDRimagesatvaryingexposurelevels,therebyallowinga16or32bits/color/pixel,butneedtobetonemappedtoSDRrangeofintensitylevelsontheorderof10,000to1.HDRfordisplayonstandardmonitors.Multi-exposurefusion(MEF)renderingalsofindsuseinc

5、omputergraphics,wherelightingtechniquesbypassHDRcreationbyfusinganexposurestackcalculationsareperformedoverawiderdynamicrange.ThisdirectlytoSDRimagestoachieveaestheticallypleasinglumi-resultsinbettercontrastvariationtherebyleadingtoahighernanceandcolordistributions.ManyHDRa

6、ndMEFdatabaseshavearelativelysmallnumberofimagesandhumanopiniondegreeofdetailpreservation.However,inordertovisualizescores,obtainedunderstringentlycontrolledconditions,therebytheseimagesonstandarddisplaydevicesdesignedforSDRlimitingrealisticviewing.Moreover,manyofthesedatab

7、asesimages,theymustbetonemappedtoSDR[2].Inadditiontoareintendedtocomparetone-mappingalgorithms,ratherthantone-mappedSDRimages,imagesarealsocreatedbymulti-beingspecializedfordevelopingandcomparingimagequalityexposurefusion,whereastackofSDRimagestakenatvaryingassessment(IQA)m

8、odels.Toovercomethesechallenges,weconductedamassivelycrowdsourcedonlinesubjectives

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。