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

Paper number 648 Bayesian structure from motion

Paper number 648 Bayesian structure from motion

Figure3:

Weperturbthehotelsequencebyreplacing5%ofthedata

pointswithdrawsfromauniformdistributionintheimageplane.TheBayesianmethod,startedasinsection3.1.1,easilydiscountsthesenoisepoints;the gureshowsthesameframesinthesequenceasin gure2,uncroppedtoshowthenoisebutwithasamplereconstructionindicatedusingthesamenotationasthat gure.

Paper number 648 Bayesian structure from motion

Figure4:Left:thelog-posteriorforeachof100samplestakenafterthe

samplerhasburnt-in(blue)comparedwiththelog-likelihoodofthestartpoint(red),forthehoteldataset.Right:thesameplot,butnowforthehoteldatasetwith5%ofthepointsreplacedwithadrawfromauniformdistributionontheimageplane.Thesmallerlog-posteriorisexplainedbythelargernumberofpointsthatmustbeassignedtonoise.

thesedataarearrangedintoamatrixofmeasure-mentswhichmustfactoras,whererep-resentsthecamerapositionsandrepresentspointposi-tions.Anaf netransformisdeterminedsuchthatminimisesasetofconstraintsassociatedwithacamera,andthenrepresentsstructure.Inpractice,factorisa-tionisachievedusingasingularvaluedecomposition.ThisisamaximumlikelihoodmethodifanisotropicGaussianerrormodelisadopted;forananisotropicGaussianerrormodel,see[18].Theformalismhasbeenappliedtovari-ouscameramodels[21,24,26];missingdatapointscanbeinterpolatedfromknownpoints[24];methodsformotionsegmentationexist[4];andmethodsforlinesandsimilarprimitivesareknown[18].Therearenoiseestimatesforre-coveredstructure[18],buttheseassumethaterrorsintheestimatesofstructureareindependent.

Maximumlikelihoodmethodsignorethevalueoftheprior.Forstructurefrommotion,thismeansthemethoddoesnotallowanypayoffbetweenmodelerror—theex-tenttowhichtherecoveredmodelviolatestherequiredsetofcameraconstraints—andmeasurementerror—theextenttowhichmodelpredictionscorrespondtodataob-servations.Thismeansthatthemodelcannotbeusedtoidentifymeasurementproblemsandsoissubjecttorecon-structionerrorscausedbyincorporatingerroneousmea-surements.Thisisasigni cantdif culty.Serioustrackererrorsleadtowildlyinaccuratereconstructions,becausethesingularvaluedecompositionofamatrixcanchangesharplywithsmallchangesintheentries.

3Bayesianstructurefrommotion

ItisusefultothinkofBayesianmodelsasgenerativemodels(e.g.[14]).Inagenerativestructurefrommotionmodel,andaredrawnfromappropriatepriors.Thenisobtainedbyaddingnoiseto.Weassumethatnoiseisobtainedfromamixturemodel;withsomelargeproba-

上一页第3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