文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > Paper number 648 Bayesian structure from motion
侵权投诉

Paper number 648 Bayesian structure from motion

Abstract:We formulate structure from motion as a Bayesian inference problem, and use a Markov chain Monte Carlo sampler to sample the posterior on this problem. This results in a method that can identify both small and large tracker errors, and yields reco

Papernumber648

Bayesianstructurefrommotion

D.A.Forsyth

S.Ioffe

ComputerScienceDivision

U.C.BerkeleyBerkeley,CA94720

daf,ioffe,haddon@cs.berkeley.edu

J.Haddon

Abstract:WeformulatestructurefrommotionasaBayesianinferenceproblem,anduseaMarkovchainMonteCarlosamplertosampletheposterioronthisprob-lem.Thisresultsinamethodthatcanidentifybothsmallandlargetrackererrors,andyieldsreconstructionsthatarestableinthepresenceoftheseerrors.Furthermore,themethodgivesdetailedinformationontherangeofambigu-itiesinstructuregivenaparticulardataset,andrequiresnospecialgeometricformulationtocopewithdegeneratesituations.Motionsegmentationisobtainedbyalayerofdiscretevariablesassociatingapointwithanobject.Wedemonstrateasamplerthatsuccessfullysamplesanap-proximationtothemarginalonthisdomain,producingarelativelyunambiguoussegmentation.

1Introduction

Figure1:

TheBayesianphilosophyisthatallinformationaboutamodeliscapturedbyaposteriordistributionobtainedusingBayes’rule:

posterior

Redpointsshowanoverheadviewofasinglesampleof

the3Dreconstructionobtainedusing40framesof80pointsinthehotelsequence,rotatedbyhandtoshowtheright-angledstructureinthemodelindicatingthatthestructureisqualitativelycorrect;thecloudofgreenpointsaresamplesofthepositionofasinglepoint,scaledby1000toshowthe(verysmall)uncertaintyavailableinasinglepointmeasurement.

likelihood

prior

joint

wherethelikelihoodisobservationsworldandtheprior,worldistheprobabilitydensityofthestateoftheworldintheabsenceofobservations.Whencomputa-tionaldif cultiescanbesidestepped,theBayesianphiloso-phyleadstoexcellentandeffectiveuseofdata(e.g.[3,8]),butcomputingwithposteriorsisdif cult.

Astandardmethodistorepresenttheposteriorbydraw-ingalargenumberofsamplesfromthatdistribution.Forexample,ifwewishedtodecidewhetherto ghtor ee,wewoulddrawsamplesfromtheposteriorandestimateexpectedutilitiesforeachdecision(assumsoftheutilities

Paper number 648 Bayesian structure from motion

overthesamples)andthenchoosethebest.Samplingalgo-rithmsaremoregeneralthanrandomsearchforMAPin-terpretationspreciselybecausetheresultsgiveanapprox-imaterepresentationoftheentireposterior.InthispaperweuseasamplingmethodtocomputeaBayesiansolutiontoastructurefrommotionproblem.Themethodisabletoidentifyanddiscounttrackererrorsthatsigni cantlyaf-fecttheperformanceoffactorisationmethods.Thesam-plerworksonanextremelylargedomain.

Notation:wewriteforavector,whose’thcompo-nentisandforamatrixwhose,’thcomponentis

.Samplerjargonthatmaybeunfamiliarisshownin

第1页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email