返回
[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost

[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost

ID：40702319

大小：2.16 MB

页数：14页

时间：2019-08-06

[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost_第1页
[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost_第2页
[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost_第3页
[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost_第4页
[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost_第5页
资源描述：

《[ECCV 2012] Metric Learning for Large Scale Image Classification Generalizing to New Classes at Near-Zero Cost》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、Authormanuscript,publishedin"ECCV-EuropeanConferenceonComputerVision(2012)"MetricLearningforLargeScaleImageClassification:GeneralizingtoNewClassesatNear-ZeroCostThomasMensink12,JakobVerbeek1,FlorentPerronnin2,andGabrielaCsurka21LEAR,INRIAGrenoble,655Avenuedel’Europe,38330Montbonnot,Franceemai

2、l:firstname.lastname@inria.fr2TVPA,XeroxResearchCentreEurope,6chemindeMaupertuis,38240Meylan,Franceemail:firstname.lastname@xrce.xerox.comAbstract.Weareinterestedinlarge-scaleimageclassificationandespeciallyinthesettingwhereimagescorrespondingtoneworexistingclassesarecon-tinuouslyaddedtothetr

3、ainingset.Ourgoalistodeviseclassifierswhichcanincorporatesuchimagesandclasseson-the-flyat(near)zerocost.Wecastthisproblemintooneoflearningametricwhichissharedacrossallclassesandex-plorek-nearestneighbor(k-NN)andnearestclassmean(NCM)classifiers.WelearnmetricsontheImageNet2010challengedataset,whi

4、chcontainsmorethan1.2Mtrainingimagesof1Kclasses.Surprisingly,theNCMclassifiercomparesfavorablytothemoreflexiblek-NNclassifier,andhascomparableperformancetolinearSVMs.Wealsostudythegeneralizationperformance,amongothersbyusingthelearnedmetricontheImageNet-10Kdataset,andweobtaincompeti-tiveperform

5、ance.Finally,weexplorezero-shotclassification,andshowhowthezero-shotmodelcanbecombinedveryeffectivelywithsmalltrainingdatasets.1IntroductionInthelastdecadewehavewitnessedanexplosionintheamountofimagesandvideosthataredigitallyavailable,e.g.inbroadcastingarchivesorsocialmediasharingweb-sites.On

6、lyasmallfractionofthisdataisconsistentlyannotatedandthusscalablemeth-odsareneededforannotationandretrievaltoefficientlyaccessthishugevolumeofdata.Thisneedhasbeenrecognizedinthecomputervisionresearchcommunityandlarge-hal-00722313,version1-1Aug2012scalemethodshavebecomeanactivetopicofresearchin

7、recentyears,seeamongoth-ers[1,2,3,4,5,6,7,8,9,10].TheintroductionoftheImageNetdataset[1],whichcontainsmorethan14Mmanuallylabeledimagesof22Kclasses,hasprovidedanimportantbenchmarkforlarge-scaleimageclassificationandannotationalgorithms.Inthispaperwef

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。