资源描述：
《Large Margin Coupled Mapping for Low Resolution Face Recognition》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、LargeMarginCoupledMappingforLowResolutionFaceRecognition1(&)1,3,412JiaqiZhang,ZhenhuaGuo,XiuLi,andYoubinChen1GraduateSchoolatShenzhen,TsinghuaUniversity,Shenzhen,Chinazhang-jq13@mails.tsinghua.edu.cn2HuazhongUniversityofScienceandTechnology,Wuhan,China3KeyLaboratoryofMeasurementandControlofCo
2、mplexSystemsofEngineering,MinistryofEducation,SoutheastUniversity,Nanjing,China4KeyLaboratoryofIntelligentPerceptionandSystemsforHigh-DimensionalInformation,MinistryofEducation,NanjingUniversityofScienceandTechnology,Nanjing,ChinaAbstract.Traditionalfacerecognitionalgorithmscanachievesignifica
3、ntperformanceunderwell-controlledenvironments.However,thesealgorithmsperformpoorlywhentheresolutionofthefaceimagesvaries.Atwo-stepframeworkisproposedtosolvetheresolutionproblemthroughadoptingsuper-resolution(SR)andperformingfacerecognitiononthesuper-resolvedfaceimages.However,suchmethodusuall
4、yhaspoorperformanceonrecog-nitiontasksasSRfocusesmoreonvisualenhancement,ratherthanclassificationaccuracy.Recently,CoupledMapping(CM)hasbeenintroducedintofacerecognitionframeworkacrossdifferentresolutions,whichlearnsacommonfeaturesubspaceforbothhigh-resolution(HR)andlow-resolution(LR)faceimage
5、s.Inthispaper,inspiredbymaximummarginprojection,weproposeLargeMarginCoupledMapping(LMCM)algorithm,whichlearnsprojectionstomaximizethemarginbetweendistanceofbetween-classsubjectsanddistanceofwithin-classonesinthecommonspace.ExperimentsonpublicFERETandSCfacedatabasesdemonstratethatLMCMiseffecti
6、veforlow-resolutionfacerecognition.Keywords:CoupledMappingLow-resolutionfacerecognitionLargeMarginCoupledMappingFERETSCface1IntroductionAgreatnumberofachievementshavebeenmadeintheareaofautomaticfacerecog-nitionduringlastdecades,especiallyunderwell-controlledcircumstances.However,theperfor
7、manceoffacerecognitionsysteminrealworldalwaysdegradesdramaticallywhenthequalityofinputfaceimagesbecomespoor,suchaslow-resolution.Thisisaspecificconcerninsurveillanceenvironmentwherethetargetisfarfromthesensor,resultinginlow-resolutionfaceimage
