Efcient Head Pose Estimation with Gabor Wavelet Networks
Ef cientHeadPoseEstimationwithGabor
WaveletNetworks
VolkerKrueger,SvenBrunsandGeraldSommer
ComputerScienceInstitute,Christian-AlbrechtsUniversityKiel
Preußerstr.1-9,24105Kiel,Germany
Tel:++49-431-560496,FAX:++49-431-560481
email:vok@ks.informatik.uni-kiel.de
Abstract
Inthisarticlewewanttointroduce rsttheGaborwaveletnetworkasamodelbasedapproachforaneffectiveandef cientobjectrepresentation.TheGa-borwaveletnetworkhasseveraladvantagessuchasinvariancetosomede-greewithrespecttotranslation,rotationanddilation.Furthermore,theuseofGabor ltersensuredthatgeometricalandtexturalobjectfeaturesareen-coded.ThefeasibilityoftheGabor ltersasamodelforlocalobjectfea-turesensuresaconsiderabledatareductionwhileatthesametimeallowinganydesiredprecisionoftheobjectrepresentationrangingfromasparsetoaphoto-realisticrepresentation.InthesecondpartofthepaperwewillpresentanapproachfortheestimationofaheadposethatisbasedontheGaborwaveletnetworks.
1Introduction
Recently,model-basedapproachesfortherecognitionandtheinterpretationofimagesofvariableobjects,likethebunchgraphapproach,PCA,eigenfacesandactiveappearancemodels,havereceivedconsiderableinterest[23;14;4;6].Theseapproachesachievegoodresultsbecausesolutionsareconstrainedtobevalidinstancesofamodel.Intheseapproaches,theterm“model-based”isunderstoodinthesensethatasetoftrainingob-jectsisgivenintheformofgrayvaluepixelimageswhilethemodel“learns”thevariancesofthegrayvalues(PCA,eigenfaces)or,respectively,theGabor lterresponses(bunchgraph).Withthis,modelknowledgeisgivenbythevariancesofpixelgrayvalues,whichmeansthattheactualknowledgerepresentationisgivenonapixelbasis,thatisindepen-dentfromtheobjectsthemselves.
InthisworkwewanttointroduceanovelapproachforobjectrepresentationthatisbasedonGaborWaveletNetworks.GaborWaveletNetworks(GWN)arecombiningtheadvantagesofRBFnetworkswiththeadvantagesofGaborwavelets:GWNsrepresentanobjectasalinearcombinationofGaborwaveletswheretheparametersofeachoftheGaborfunctions(suchasorientationandpositionandscale)areoptimizedtore ecttheparticularlocalimagestructure.Gaborwaveletnetworkshaveseveraladvantages:
1.Bytheirverynature,Gaborwaveletnetworksareinvarianttosomedegreetoaf nedeformationsandhomogeneousilluminationchanges,
Gabor Wavelet Networks f... 14页 免费 Head Pose Estimation for... 暂无评价 10页 免费 Efcient Head Pose Estima... 10页 免费 Real Time Head Pose ...
Gabor Wavelet Networks... 14页 免费 Image representation b... 8页 免费 3D wavelet-based multi... 17页 4下载券 Efcient Head Pose Esti... 10页 ...
Efcient Head Pose Estima... 10页 免费 Junior Tracking Facial F... 暂无... Gabor wavelet networks for object representation Abstract. In this article ...
Gabor Wavelet Networks f... 14页 免费 Gabor Wavelet Selection ... 暂无... Efcient Head Pose Estima... 10页 免费 Junior Tracking Facial F... 暂无...
Efcient Head Pose Esti... 10页 免费 Efficient head pose es... 暂无...B. Friction Estimation with Wavelet Networks Only the friction in joint 1 ...
Gabor Wavelet Networks f... 14页 免费 Efcient Head Pose Estima... 10... Trasformata Wavelet e Trasformata di Gabor a confronto google学术搜索google...
暂无评价 31页 免费 Wavelet neural networks ... 13页 1财富值... 10页 免费 Efcient Head Pose Estima... 10页 免费搜你所想,读你所爱 ...
Efcient Head Pose Estima... 10页 免费 ... Statistical Pose Estimation with Local Dependencies...They can be extracted from wavelet decompositions ...
Minimax wavelet estimati... 暂无评价 30页 免费 Minimax estimation with ..... Wavelet networks for est... 6页 5财富值 GABOR WAVELET BASED P... 6...
GABOR WAVELET BASED P... 6页 免费 Model based pose estimat... 13页 ... C-IMPLEMENTATION OF A NEW PROCEDURE FOR GUIDE MARK BASED POSE ESTIMATION...
热门文档