a_general_framework_for_object_detection

a_general_framework_for_object_detection

ID:39910186

大小:1.35 MB

页数:8页

时间:2019-07-14

a_general_framework_for_object_detection_第1页
a_general_framework_for_object_detection_第2页
a_general_framework_for_object_detection_第3页
a_general_framework_for_object_detection_第4页
a_general_framework_for_object_detection_第5页
资源描述:

《a_general_framework_for_object_detection》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、AGeneralFrameworkforObjectDetectionConstantineP.PapageorgiouMichaelOrenTomasoPoggioCenterforBiologicalandComputationalLearningArtificialIntelligenceLaboratoryMITCambridge,MA02139{cpapa,oren,tp}@ai.mit,eduAbstractintheimage;MAPormaximumlikelihoodmethod

2、swillnotworksincetheclassificationofeachpatternThispaperpresentsageneraltrainableframeworkinanimageisdoneindependently.Thispaperalsoforobjectdetectioninstaticimagesofclutteredscenes.introducesanextensionthatusesmotioncuestoim-Thedetectiontechniquewede

3、velopisbasedonaprovedetectionaccuracyovervideosequences.Thiswaveletrepresentationofanobjectclassderivedfromamotionmoduleisageneralonethatcanbeusedwithstatisticalanalysisoftheclassinstances.Bylearningmanydetectionalgorithmsanddoesnotcompromiseanobjectc

4、lassintermsofasubsetofanovercompletetheabilityofthesystemtodetectnon-movingobjects.dictionaryofwaveletbasisfunctions,wederiveacom-Initialworkonthedetectionofrigidobjectsinpactrepresentationofanobjectclasswhichisusedasstaticimages,suchasstreetsignsorfa

5、ces,Betke&aninputtoasupporivectormachineclassifier.ThisMakris[l],Yuille,et.al.[2l],usedtemplatematch-representationovercomesboththeproblemofin-classingapproacheswithasetofrigidtemplatesorhand-variabilityandprovidesalowfalsedetectionrateincraftedparame

6、terizedcurves.Theseapproachesareunconstrainedenvironments.difficulttoextendtomorecomplexobjectssuchasWedemonstratethecapabilitiesofthetechniqueinpeople,sincetheyinvolveasignificantamountofpriortwodomainswhoseinherentinformationcontentdif-informationan

7、ddomainknowledge.Inrecentre-ferssignificantly.Thefirstsystemisfacedetectionsearch,morecloselyrelatedtooursystem,thedetec-andthesecondisthedomainofpeoplewhich,incon-tionproblemissolvedusinglearning-basedtechniquestrasttofaces,varygreatlyincolor,texture

8、,andpat-thataredatadriven.ThisapproachwasusedbySungterns.Unlikepreviousapproaches,thissystemlearns&Poggio[lG]andVaillant,etal.[l8]forthedetectionfromexamplesanddoesnotrelyonanyapriori(hand-offrontalfacesinclutteredscenes,withsimilararcbi-craft

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。
相关文章
更多
相关标签