预览加载中，请您耐心等待几秒...

基于Gabor相位和局部二值模式的AAM纹理表示.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于Gabor相位和局部二值模式的AAM纹理表示 Abstract: ActiveAppearanceModels(AAMs)havebeenwidelyusedinfaceandobjectrecognitionduetotheirabilitytomodelthevariationsinfacialappearance.AkeycomponentofAAMsisthetexturerepresentation,whicheffectivelycapturesthevariationsinthefacialtextures.Inthispaper,weproposeanovelAAMtexturerepresentationbasedonGaborphaseandLocalBinaryPatterns(LBP),whichcanrobustlyrepresentthefacialtextureandhandlevariousilluminationandposevariations.ExperimentalresultsontheHELENandLFPWdatasetsdemonstratethattheproposedmethodachievessuperiorperformancecomparedtostate-of-the-artmethods. Introduction: Withtherapiddevelopmentofcomputervisionandmachinelearning,AAMshavebecomeapopulartoolformodelingandanalyzingtheshapeandappearanceoffacesandotherobjects.AAMsareabletoeffectivelycapturethevariationsinfacialappearancecausedbyfactorssuchaspose,lighting,andexpression.However,thekeytothesuccessofAAMsliesinthetexturerepresentationmethodtheyemploy.TexturerepresentationisacriticalcomponentofAAMsbecauseitisusedtomodelthevariationsinfacialtexturesthatarerelatedtofacialappearance. Inrecentyears,varioustexturerepresentationmethodshavebeenproposedforAAMs,includinglocalbinarypatterns(LBP)andGaborfilters.However,thesemethodshavesomelimitationsinmodelingthecomplexvariationsinfacialappearance.Forexample,LBPissensitivetonoiseandblur,whileGaborfiltersaresusceptibletoilluminationchangesandrotation.Inthispaper,weproposeanoveltexturerepresentationmethodthatcombinestheadvantagesofGaborphaseandLBPtoovercometheselimitations. Methodology: TheproposedAAMtexturerepresentationmethodconsistsoftwomainsteps:featureextractionandfeatureencoding.Inthefeatureextractionstep,weuseGaborfilterstoextractthephaseinformationofthetexture.Gaborfiltersareabletomodelthevariationsinthefrequencyandorientationofthetexture,whichareimportantforcapturingthevariationsinfacialtextures.ThephaseinformationobtainedfromtheGaborfiltersisthenusedtorepresentthetexture. Inthefeatureencodingstep,weuselocalbinarypatterns(LBP)toencodethetextur