資源描述：
《2014-ICLR-OverFeat_ Integrated Recognition, Localization and Detection using Convolutional Networks》由會員上傳分享，免費在線閱讀，更多相關內容在學術論文-天天文庫。
1、OverFeat:IntegratedRecognition,LocalizationandDetectionusingConvolutionalNetworksPierreSermanetDavidEigenXiangZhangMichaelMathieuRobFergusYannLeCunCourantInstituteofMathematicalSciences,NewYorkUniversity719Broadway,12thFloor,NewYork,NY10003sermanet,deigen,xiang,mat
2、hieu,fergus,yann@cs.nyu.eduAbstractWepresentanintegratedframeworkforusingConvolutionalNetworksforclassi-?cation,localizationanddetection.Weshowhowamultiscaleandslidingwindowapproachcanbeef?cientlyimplementedwithinaConvNet.Wealsointroduceanoveldeeplearningapproachto
3、localizationbylearningtopredictobjectbound-aries.Boundingboxesarethenaccumulatedratherthansuppressedinordertoincreasedetectioncon?dence.Weshowthatdifferenttaskscanbelearnedsimul-taneouslyusingasinglesharednetwork.Thisintegratedframeworkisthewinnerofthelocalizationt
4、askoftheImageNetLargeScaleVisualRecognitionChallenge2013(ILSVRC2013)andobtainedverycompetitiveresultsforthedetectionandclassi?cationstasks.Inpost-competitionwork,weestablishanewstateoftheartforthedetectiontask.Finally,wereleaseafeatureextractorfromourbestmodelcalle
5、dOverFeat.1IntroductionRecognizingthecategoryofthedominantobjectinanimageisataskstowhichConvolutionalarXiv:1312.6229v4[cs.CV]24Feb2014Networks(ConvNets)[17]havebeenappliedformanyyears,whethertheobjectswerehandwrittencharacters[16],housenumbers[24],texturelesstoys[1
6、8],traf?csigns[3,26],objectsfromtheCaltech-101dataset[14],orobjectsfromthe1000-categoryImageNetdataset[15].TheaccuracyofConvNetsonsmalldatasetssuchasCaltech-101,whiledecent,hasnotbeenrecord-breaking.However,theadventoflargerdatasetshasenabledConvNetstosigni?cantlya
7、dvancethestateoftheartondatasetssuchasthe1000-categoryImageNet[5].ThemainadvantageofConvNetsformanysuchtasksisthattheentiresystemistrainedendtoend,fromrawpixelstoultimatecategories,therebyalleviatingtherequirementtomanuallydesignasuitablefeatureextractor.Themaindis
8、advantageistheirravenousappetiteforlabeledtrainingsamples.Themainpointofthispaperistoshowthattrainingaconvolutionalnetworktosimultaneouslyclassif
