文档文库
手机版
投诉建议
热门搜索： 心得体会 演讲稿 思想汇报
首页 > CVPR_2014_Pocket_Guide_Final

CVPR_2014_Pocket_Guide_Final

发布时间：   来源：文档文库   
小 中 大
字号：
手机查看







MessagefromtheGeneralandProgramChairs
WelcometoColumbus,Ohioandthe27thIEEEConferenceonComputerVisionandPatternRecognition(CVPR.Inadditiontothemainfour-dayprogramoforalandposterpresentations,keynotetalks,demos,exhibitions,andsocialfunctions,CVPR2014hasanumberofco-locatedevents,including19workshopsand15tutorials.
Thisyear,wereceived1807validsubmissionstothemainconference,ofwhich1778werefullyreviewed.(Theotherswereeitheradministrativelyrejectedfortechnicalorethicalreasonsorwithdrawnbeforereview.Toselectpapersfromthesesubmissions,weinvited55highlyregardedresearcherstoactasAreaChairs(ACs.ACswereselectedtoincludeabroadrangeofgeographicallocationswhileprovidingbalanceinbothgenderandseniority.NomorethanoneACperinstitutionwasselected.ThenumbersofseniorandjuniorACswereapproximatelyequaltoallowthepairingofmoreexperiencedACswithlessexperiencedonesduringpaperdiscussions.
Werecruitedanexpertteamof981experiencedreviewersfromthebroaderCV&PRcommunity.Ourcriteriatoselectreviewerswereaprovenpublicationrecordinthetopcomputervisionconferencesandjournalsandacompleted(orclosetocompleteddoctoraldegree.TheoriginallistofreviewerswasslightlyincreasedbyadditionalreviewersrecommendedbytheACstoaddexpertiseforpaperswhereappropriatereviewerswerenotavailable.
Asinpreviousyears,weusedtheCMTconferencemanagementserviceprovidedbyMicrosoftResearchtomanagethesubmissionandselectionofpapersfrombeginningtoend.Also,forthefirsttime,weusedtheiThenticatesoftwaretodetectpotentialcasesofplagiarism.
Afterthesubmissiondeadline,theProgramChairsdistributedthepaperstotheACswithhelpfromtheautomatedTorontoPaperMatchingSystem(TPMS,developedbyCharlinandZemel[ICML2013].TPMSsuggestsmatchesbetweenpapersandACsbasedonthePDFfilesofsubmittedmanuscriptsandrepresentativepublicationsbyeachAC;forCVPR2014wehadtwoTechnologyChairswhowerein
1

chargeoftheinterfacewithTPMS.TheACs,inturn,usedTPMStohelpthemdeterminethepotentialreviewersforeachoftheirassignedpapers.ACssuggested5-10reviewersperpaperfromwhichCMTautomaticallyselectedthreenon-conflictedreviewersperpaper.Finally,manualadjustmentsweremadebytheACsandProgramChairstoachievebettermatchesbetweenthepapersandreviewers.Asaresult,eachreviewerwasassignedamaximumof9papers,withamedianloadof7papers.
TwokeyinnovationstothereviewformforCVPR2014weretoincludean‘oral/poster’ratingandtoeliminatetheborderlinerating.Reviewersweregivensixweekstocompletetheirreviews,atwhichtimetheACssteppedbackintovetthereviewsforquality(initiatingdiscussions,wherenecessaryandwriteaninitialconsolidationreportofthereviewsbeforereleasingthemtotheauthors,whoweregivenoneweektowriterebuttals.Aftercollectingtherebuttals,theACsfinishedtheirpre-meetingwork,i.e.,consolidatingthereviewsandauthorrebuttals,initiatingdiscussionsamongthereviewersforclarification,andmakingrecommendationsfordecisionsonpapers.Reviewerswereaskedtocarefullyreadandrespondtotheauthors’rebuttals.TheProgramChairsandtheACsstrovetoensurethateverypapereligibleforfullreviewreceivedatleastthreehighqualityreviewsandathoughtfulfinalconsolidationreportthatsummarizedthereviewerdiscussion.
Everypaper,review,andauthorrebuttalwasevaluatedbyaprimaryAC,whomadeapreliminaryrecommendationonthepaper.AllpapersthatwererecommendedforrejectionwerethenreviewedbyasecondaryAC.PapersthatwerenotconsideredacceptablebybothACswererejectedbeforetheACmeeting,andtheACwroteaconsolidationreportexplainingthedecision.The984papersthatpassedthisreviewwerediscussedattheACmeetingheldinFebruaryattheUniversityofMaryland.
AttheACmeeting,theACsweredividedinto18panelsofthreememberseach,withnoconflictsbetweentheACsandpapersassociatedwitheach

MessagefromtheGeneralandProgramChairs
panel.TheProgramChairsmonitoredthepaneldiscussionsandworkedhardtomaintainconsistencyamongthepanels.DecisionstoacceptorrejectapaperweremadebythethreeACsworkingtogetherand,asneeded,byinputfromadditionalnon-conflictedACs.Bytheendofthediscussion,theACswereaskedtoproducedetailedconsolidationreportstojustifyalltheirdecisions.Thepanelsselected540papersforpresentationatthemainconference(a29.88%acceptancerate.Paperswereselectedonmerit,nottofulfilanyquotas.Panelsofatleastsixnon-conflictingACswereassembledtoevaluateacceptedpapers.Thepanelsselected104ofthesepapers(5.75%tobepresentedasoralsattheconference.
TheProgramChairsandGeneralChairsdidnotsubmitanypaperstoCVPR2014,allowingthemtoworkwithoutanydirectconflictsthroughoutthereviewprocess.(TheGeneralChairswerealsonotpresentattheACMeeting.Additionally,therespectiveACswereexcludedfromanydecisionsassociatedwithpapersfromtheirresearchgroups,affiliatedinstitutionsorcollaborators.Thedouble-blindnatureoftheCVPRreviewprocesswasthusstrictlymaintainedthroughout.
TheoverallacceptanceratewascomparabletothoseofpreviousCVPRs.However,thepercentageoforalpresentationswassignificantlyincreasedfromrecentyears,from3.3%lastyearto5.75%thisyear.ThiswasadeliberatedecisionbasedonextensiveconsultationwithmembersoftheCVPRcommunity.ThemainconferenceatCVPRwasscheduledasathree-dayeventmanyyearsago,whenthenumberofsubmissionswasfewerthanhalfofthatseeninrecentyears.Asthenumberofsubmissionshasgrown,thepercentageoforalpresentationshasdecreased.Atfirst,thisincreasedthequalityoforalpresentations,butiteventuallybecamefrustratingforreviewers,ACs,andauthorsthatmanyexcellentpaperscouldnotbeselectedfororalpresentation.
Themainconferencethisyearhasbeenexpandedfromthreetofourdays,attheexpenseoflosingaworkshop/tutorialday(tomaintainthesameoverall
2

conferencelength.Posterspotlightshavebeenmovedtovideospotlights(availablefororalpresentersaswell,whichareavailableonlinebefore,duringandaftertheconference.Thishasallowedforasignificantincreaseinoralpresentationstoprovidemoreauthorsanopportunitytopresenttheirresearchtoabroaderaudience.
AnotherinnovationofCVPR2014isthecreationofaPAMI/IJCVjournalsession.ThegoalofthissessionistoexposetheCVPRcommunitytohigh-impactjournalpublicationsthathaveappearedinrecentyearsandwerenotpreviouslypresentedatamajorcomputervisionconference.Sixrecenthigh-impactjournalpaperswereselectedbyacommitteeconsistingoftwoformerEditors-in-ChiefofPAMI.Thesepaperswereselectedfromapoolof20+papersnominatedbyAssociateEditorsofPAMIandIJCV.Themainmeetingalsoincludestwoinvitedspeakers,Dr.DorisTsaoandDr.StéphaneMallat.
TheproceedingsofCVPR2014arebeingprovidedonUSBdrivesattheconference.AllpapersinthemainconferenceandassociatedworkshopswillbemadeavailablethroughtheIEEEComputerSocietyDigitalLibraryandthroughIEEEXplore.
WewishtothankallmembersoftheOrganizingCommittee,theAreaChairs,reviewers,authors,andtheCMTandTPMSteamsfortheimmenseamountofhardworkandprofessionalismthathasgoneintomakingCVPR2014afirst-rateconference.OurthanksalsogototheorganizersofpreviousCVPRsfortheirhelpfuladviceandsupport.Wearegratefultothesponsorsaswell,andwearehappytoreportthatCVPR2014hasseenasignificantincreaseinindustrialsupport,whichisfurtherevidenceoftherelevanceandimportanceofthiscommunity.
Finally,wewishalltheattendeesahighlystimulating,informative,andenjoyableconference.
SvenDickinson,DimitriMetaxas,andMatthewTurk,GeneralCo-Chairs
RonenBasri,CorneliaFermuller,AleixMartinez,andRenéVidal,ProgramCo-Chairs

OrganizingCommittee,AreaChairs,&OutstandingReviewers
GeneralChairs:
ProgramChairs:
WorkshopsChairs:
TutorialsChairs:
FinanceChairs:
WebsiteChairs:
LourdesAgapitoTalArbel
KobusBarnardTamaraBergHorstBischofDanielCremersRitaCucchiaraKostasDaniilidisPiotrDollar
AhmedElgammalPaoloFavaro
CVPR2014OrganizingCommittee
SvenDickinsonPublicationsChairs:DimitriMetaxasMatthewTurkDemos/ExhibitionsChair:RonenBasriCorporateRelationsChairs:CorneliaFermullerAleixMartinezRenéVidal
DoctoralConsortiumChairs:
JimDavisMargritBetke
PublicityChair:
KristenGrauman
VideoProceedingsChair:
RaquelUrtasun
StudentActivitiesChair:
TerryBoult
LocalArrangementsChair:OctaviaCamps
TechnologyChairs:RyanFarrell
GaryHuang
LogisticsAdvisor:YaserYacoobLeoGrady
EdwinHancockTalHassnerDavidJacobsHuiJi
FredericJurieFredrikKahl
IoannisKakadiarisJanaKosecka
ChristophLampertSvetlanaLazebnik
CVPR2014AreaChairsErikLearned-MillerZhouchenLinHaibinLingSimonLuceyYiMaJiriMatas
GerardMedioniGregMoriNikosParagiosRobertPlessPetiaRadeva
EricMortensenSanjaFidlerMichaelBrownKariPulliGangHuaJensRittscher
PhilipposMordohaiRaquelUrtasunKristinDana
IoannisKakadiarisJasonCorsoBrianKulisYaserSheikhJingyiYuGingerBoult
RaquelUrtasunManikVarma
NunoVasconcelosOlgaVekslerBabaVemuri
DaphnaWeinshallAllenYang
Ming-HsuanYangLaurentYounesLihiZelnik-Manor
JamesRehgArunRossSudeepSarkarYoichiSato
KonradSchindlerThomasSerreRahulSukthankarSinisaTodorovicCarloTomasi
FernandodelaTorreZhuowenTu
CVPR2014OutstandingReviewers
Wearepleasedtorecognizethefollowingresearchersasinprovidingdetailedreviewsforthepapersassignedtothem."OutstandingReviewersforCVPR2014".ThesereviewersThesereviewerswereidentifiedbyoneormoreoftheCVPRwereselectedfromalmost1000reviewersfortheirhardworkAreaChairs,whofoundtheirreviewsofhighquality.ElliAngelopoulouJoaoBarreto
OhadBen-ShaharMosheBen-ezraMargritBetkeRossBeveridgeThomasBroxOctaviaCampsAntoniChanTsung-hanChanBobCollins
JasonCorsoJamesCrowleyAlessioDelBue
KonstantinosDerpanisJan-MichaelFrahmFabioGalassoPeterGehelerChristopherGeyerMichalHavlenaXumingHeAdrianHilton
AnthonyHoogsOmarJavedHaoJiangRolandKwitt
Jean-FrancoisLalondeIvanLaptevSubhransuMajiTimMarks
YasuyukiMatsushitaPeymanMilanfar
FrancescMoreno-Noguer
3

EricMortensenKoNishinoBjornOmmerDeviParikhNikhilRasiwasiaStefanRoth
MohammadSaberianAlbertSalahRadimSara
RainerStiefelhagenYusukeSugano
DavidSuterYuichiTaguchiRobbyTanRobertoTronStefanWalkJingdongWangLiorWolfJohnWrightLarryZitnick

Monday,June23
Monday,June23
0700–1700Registration(ExhibitHallCLobby0730–0830Breakfast(ExhibitHallC1200–1330Lunch(ExhibitHallC

PerceptualOrganization
Organizers:MichaelMaireStellaYuLocation:C115Schedule:FullDay
0830Welcome
S1:Segmentation&Video(0830–1010
0830Movingvs.StaticObjectsinVideoSegmentation,
ThomasBrox(Univ.ofFreiburg0850CompositeStatisticalLearningandInferencein
SemanticandVideoSegmentation,FuxinLi(GeorgiaInstituteofTechnology0910InteractiveLearningforPoint-CloudMotion
Segmentation,TalHassner(OpenUniv.ofIsrael0930LearningfromYouTubeVideos,RahulSukthankar
(Google0950SpotlightPresentations1000MorningBreak
S2:Representation(1030–1200
1030FromEdgestoObjects,PiotrDollar(Microsoft
Research1050PerceivingCrowds:BypassingtheBottleneckof
ConsciousVision,DavidWhitney(UCBerkeley1130Hierarchy,Reasoning,andRepresentationLearning,
YannLeCun(NewYorkUniv.

Workshops
1200LunchBreak(ExhibitHallCS3:Activity(1330–1500
1330LiveCountingofRepetitiveActionsviaaConvolutional
DeepNeuralNetworkTrainedonUnrealisticSyntheticData,LiorWolf(TelAvivUniv.1350Weakly-SupervisedLearningofActions,IvanLaptev
(INRIA1410RecognizingHumanActivities:Interactions,Groups,
andContext,GregMori(SimonFraserUniv.1430VideoSceneSegmentationandRecognitionby
Location-IndependentActivityClasses,AnthonyHoogs(Kitware1450SpotlightPresentations1500AfternoonBreak
S4:Motion,Shape,&Parts(1520–1700
1520FromShallowHollywoodMotionstoDeepAcademic
Gestures,ChristophBregler(NewYorkUniv.1540PerceptualOrganizationofMotion:EventRecognition
andthePerceptionof3DStructurefromMotion,JamesTodd(OhioStateUniv.1620Data-DrivenPerceptualInterpretationofShape,
WilliamFreeman(MIT1640DoMid-LevelPartsStillMatterintheAgeofCNNs?
LubomirBourdev(Facebook


4

Monday,June23
MobileVision
Organizers:ZhengyouZhangMarcPollefeysGangHuaMatthewTurkKariPulliRajaBalaLocation:C210Schedule:FullDay
0825OpeningRemarks,GangHua
S1:MobileVisualRecognitionandSearch(0830-0930
Workshops
S3:Mobile3DModelingandOtherMobileRelated
Applications(1425-152514253DHallwayModelingUsingASingleImage,Greg
Olmschenk,ZhigangZhu1445EstimatingGazeDirectionofVehicleDriversusinga
SmartphoneCamera,Meng-CheChuang,RajaBala,EdgarA.Bernal,PeterPaul,AaronBurry1505GPSRefinementandCameraOrientationEstimation
fromaSingleImageanda2DMap,HangChu,AndrewGallagher,TsuhanChen1525AfternoonBreak
S4:OtherMobileRelatedApplications(1555-16351555FastandRobustObjectDetectionUsingVisual
Subcategories,EshedOhn-Bar,MohanM.Trivedi1615VisiononWheels:LookingatDriver,Vehicle,and
SurroundforOn-RoadManeuverAnalysis,EshedOhn-Bar,AshishTawari,SujithaMartin,MohanM.TrivediS5:Demos(1640-1730
1640Space-VariantImageDeblurringonSmartphonesusing
InertialSensors,OndřejŠindelář,FilipŠroubek,PeymanMilanfar1650Offline1000-ClassClassificationonaSmartphone,
YoshiyukiKawano,KeijiYanai1700ACompact3DCameraSuitedforMobileand
EmbeddedVisionApplications,StefanoMattoccia,IlarioMarchio,MarcoCasadio1710FastandRobustPerspectiveRectificationofDocument
ImagesonaSmartphone,Williem,ChristianSimon,SungdaeCho,InKyuPark1720CorrectingPhotometricDistortionofDocument
ImagesonaSmartphone,ChristianSimon,Williem,JihwanChoe,IlDongYun,InKyuPark1730BestPaperAwardAnnouncement(Sponsoredby
MicrosoftResearchandNvidiaResearch
0830FastTargetRecognitiononMobileDevices:Revisiting
GaussianEliminationforthetheEstimationofPlanarHomographies,OlexaBilaniuk,HamidBazargani,RobertLaganière0850CascadeofBox(CABOXFiltersforOptimalScale
SpaceApproximation,VictorFragoso,Gaurav
Srivastava,AbhishekNagar,ZhuLi,KyungmoPark,MatthewTurk0910Real-timeMobileFacialExpressionRecognition
System–ACaseStudy,MyunghoonSuk,BalakrishnanPrabhakaran0930InvitedTalk:TowardsUbiquitousEmbedded3DVisual
SensinginMobileDevices,AchintyaBhowmik(IntelCorporation1015MorningBreak
S2:MobileComputationalPhotographyandMultiview
Analysis(1045-11451045DenseViewInterpolationonMobileDevicesusing
FocalStacks,ParikshitSakurikar,P.J.Narayanan1105DynamicImageStacks,DavidE.Jacobs,OrazioGallo,
KariA.Pulli1125RobustThree-viewTriangulationDoneFast,Johan
Hedborg,AndreasRobinson,MichaelFelsberg1200LunchBreak(ExhibitHallC
1345InvitedTalk:MobileImaging:TheFutureOfThe
Image,EdwardJ.Delp(PurdueUniv.
5



Monday,June23
SceneUnderstanding
Organizers:JamesHaysDerekHoiemAdityaKhoslaJianxiongXiaoLocation:C214-215Schedule:FullDay
0830WelcomebyOrganizers
0835InvitedTalk:IrvingBiederman(Univ.ofSouthern
California0905InvitedTalk:JitendraMalik(Univ.ofCalifornia,
Berkeley0935SponsorInvitedTalk:ChangHuang(BaiduInst.of
DeepLearning0945SponsorInvitedTalk:YuanqingLin(NECLabs1000MorningBreak
1030InvitedTalk:AlexeiAEfros(Univ.ofCalifornia,
Berkeley1100InvitedTalk:AshutoshSaxena(CornellUniversity1130PosterSpotlights
1200LunchBreak(ExhibitHallC1330PosterSession1500AfternoonBreak
1530InvitedTalk:DavidForsyth(Univ.ofIllinois,Urbana-Champaign1600InvitedTalk:RobFergus(NewYorkUniv.1630SponsorInvitedTalk:RahulSukthankar(Google
Research

VisionMeetsCognition
Organizers:YibiaoZhaoLap-FaiYuBoZhengPeterBattagliaTaoGao
Location:GrandBallroom1Schedule:FullDay
0830WelcomeMessage
Workshops
0840InvitedTalk:BeyondWhatandWhere:JointSpatial,
TemporalandCausalParsingwithCommonsenseReasoning,Song-ChunZhu(UCLA0925InvitedTalk:Learning-from-Observation:From
AssemblyRobotThroughDancingHumanoid,KatsushiIkeuchi(Univ.ofTokyo0955InvitedTalk:SeeingTime'sArrow/InferringProperties
ofClothfromWatchingitMove,BillFreeman(MIT1025MorningBreak
1045InvitedTalk:CognitiveVisionanditsApplicationto
MedicalImaging,VisualSurveillance,andSpaceRobotics,DemetriTerzopoulos(UCLA1115Posters
1200LunchBreak(ExhibitHallC1300Posters
1355InvitedTalk:ModelingHumanCommon-SenseScene
Understanding,JoshTenenbaum(MIT1440InvitedTalk:Physics,Humans,andIntention:
InterpretableMachineLearningfor3DSceneUnderstanding,AshutoshSaxena(CornellUniv.1510InvitedTalk:VisionisforAgents,BenjaminKuipers
(Univ.ofMichigan1540AfternoonBreak
1600InvitedTalk:SeeingIntotheFuture,LarryZitnick
(MicrosoftResearch1630Paneldiscussion
6
1640InvitedTalk:AntonioTorralba(MassachusettsInst.of
Technology1710InvitedTalk:LarryZitnick(MicrosoftResearch1720InvitedTalk:MartialHebert(CarnegieMellon
University1750SponsorInvitedTalk:Marc’AurelioRanzato(Facebook
AIResearch1800AwardCeremony,WorkshopChairs


Monday,June23
Organizers:SekChaiBoazSuperHimanshuAroraTerranceBoultArnabDhuaMarshallTappenRajaBalaYuWang
Location:GrandBallroom2Schedule:FullDay
Workshops
VideoAnalyticsatUnitedTechnologiesResearchCenter,AlanFinn
Real-timeUrbanMeteringwithPedestrianandVehicleRecognition,AlexandreWinter,IgnacioMelladoBataller,TuanThi
CollaborativeComputerVisionR&DatKitware,ArslanBasharat,SangminOh,MattLeotta,RustyBlue,KeithFieldhouse,MattTurek,BradDavis,HeatherJames,AnthonyHoogs
TeraDeep:IntelligentVisionSystems,EugenioCulurciello
ViPanix:PanoramicVideos,FranciscoHernandez-Lopez,MarianoRivera
EnhancingConfidenceinVideoAnalytics,GaryRubin,DavidBerger
PercepTonic:WeSeeSolutions,GokselDedeoglu,SusanRossbach
Real-timeImageClassificationonMobilePhones,HarroStokman,SamirKumar,DanielFontijne,OrkdeRooijAmazonFulfillmentTechnologyComputerVision,MarshallTappen
ReKognitionAPIPlatform,MengWang,TianqiangLiu,YushanChen
ComputerVisionatEyenuk:ImageAnalysisforYourHealthandYourPhotos,SandeepBhat,ChaithanyaRamachandra,MalavikaBhaskaranand,KaushalSolankiSRIInternational:BreakthroughIdeas…Real-WorldSolutions,SekChai
VisualSearchTechnologyforAmazon,SunilRamesh,ArnabDhua,HimanshuArora1500AfternoonBreak
S4:ComputerVisioninServicesIndustry(1525–16451525InvitedTalk:AnOverviewofRetailVideoAnalytics,
QuanfuFan(IBM1605InvitedTalk:ComputerVisionwithanEyeonServices,
RajaBalaandPeterPaul(XeroxS5:PanelSession(1645–1730
1645Panel:ComputerVisionIndustryandCommunity,
Moderator:TerranceBoult(UCCS1720BeyondVIEW2014:TerranceBoult(UCCS,SekChai
(SRI
7
VisionIndustry&EntrepreneurWorkshop

0835Welcome:SekChai(SRI,HimanshuArora(A9.comS1:MobilePlatformsforComputerVision(0845–11050845InvitedTalk:MobileComputationalImaging,KariPulli
(Nvidia0925InvitedTalk:ComputerVisionApplicationsforMobile
andBeyond,MaheshRamachandran(Qualcomm1005MorningBreak
1025InvitedTalk:OpenVX:TheComputerVisionHardware
AbstractionLayer,VictorEruhimov(KhronosS2:DistinguishedSpeaker(1105–1145
1105InvitedTalk:ThePitandthePendulum:Academic
ResearchinIndustrialLabs,RahulSukthankar(GoogleS3:IndustrySession:Demos,Posters,Recruiting(1145–
15001145IndustrySessionSpotlights:Moderators:ArnabDhua
(A9.com,MarshallTappen(Amazon1210LunchBreak(ExhibitHallC
1315Demos,Posters,Recruiting
ComputerVisionforEnterpriseandPublicSafetyatMotorolaSolutions,AnkurPatel
XeroxComputerVisionforRoadwayTransportationSystems:Robust,HighYieldAutomatedLicensePlateRecognition,AaronBurry,VladimirKozitsky
XeroxComputerVisionforRoadwayTransportationSystems:AutomatedImage-BasedDetectionofFrontSeatPassengersinVehicles,AaronBurry,PeterPaul,YusufArtan,FlorentPerronnin


Monday,June23
ComputerVision&HumanComputation
Organizers:JiaDengSubhransuMajiPietroPeronaLocation:GrandBallroom3Schedule:FullDay
0920OpeningRemarks
Workshops
PerceptionBeyondtheVisibleSpectrum
Organizers:RiadI.HammoudGuoliangFanFiroozSadjadiBehzadKamgar-ParsiLocation:C110
Schedule:HalfDay—Morning
0800WelcomeMessage
0810KeynoteTalk:Multi-frameDataAssociationwith
Higher-OrderCostFunctions,ErikP.Blasch(AirForceResearchLabS1:Thermal&InfraredImaging(0845-0930
0845AThermalInfraredVideoBenchmarkforVisual
Analysis,ZhengWu,NathanFuller,DianeTheriault,MargritBetke0900LowResolutionPersonDetectionwithaMoving
ThermalInfraredCamerabyHotSpotClassification,MichaelTeutsch,ThomasMüller,MarcoHuber,JürgenBeyerer0915ImprovingPersonTrackingUsinganInexpensive
ThermalInfraredSensor,SurenKumar,TimK.Marks,MichaelJonesS2:ActivityRecognition&Surveillance(0930-10150930DriverCellPhoneUsageDetectionFromHOV/HOT
NIRImages,YusufArtan,OrhanBulan,RobertP.Loce,PeterPaul0945Ground-BasedActivityRecognitionatDistanceand
BehindWall,TaoWang,RiadHammoud,ZhigangZhu1000Multi-SourceMulti-ModalActivityRecognitionin
AerialVideoSurveillance,RiadI.Hammoud,CemS.Sahin,ErikP.Blasch,BradleyJ.Rhodes1015MorningBreak:Posterpreparation
0925InvitedTalk:BuildingLargeDatasetstoRepresentthe
World,LarryZitnick(MicrosoftResearch1005MorningBreak
1030InvitedTalk:BeyondMindlessLabeling:*Really*
LeveragingHumanstoBuildIntelligentMachines,DeviParikh(ViginiaTech1105SpotlightsandPosters1200LunchBreak(ExhibitHallC
1330InvitedTalk:QuicklyAnsweringGeneralVisual
Questions,JeffreyBigham(CMU1405KeynoteTalk:EyeWire,AGametoMaptheBrain,
SebastianSeung(PrincetonUniv.1500AfternoonBreak
1530InvitedTalk:CrowdOneShotLearning,JamesHays
(BrownUniv.1605InvitedTalk:VisipediaToolEcosystemforDataset
CurationandAnnotation,SergeBelongie(CornellTech.


8


Monday,June23
S3:PointRegistration,3DEstimation,and3D
Segmentation(1045-1130
Workshops
RegistrationofVeryLargeImages
Organizers:ArdyGoshtasbyChangShuAkihiroSugimotoJohnCampNathanNetanyahuClaudeCariouHectorErivesClarkTaylorLyubomirZagorchevMartinSatterMarcelJackowskiSimonWarfieldLocation:C110
Schedule:HalfDay—Afternoon
1300InvitedTalk:EfficientHigh-ResolutionStereo
MatchingusingLocalPlaneSweeps,SudiptaSinha(MicrosoftResearch1345AutomaticGeo-locationCorrectionofSatellite
Imagery,OzgeC.Ozcanli,YiDong,JosephL.Mundy,HelenWebb,RiadHammoud,TomVictor1410EfficientChangeDetectionforVeryLargeMotion
BlurredImages,VijayRengarajan,Abhijith
Punnappurath,A.N.Rajagopalan,GunaSeetharaman1435Poster:Non-rigidRegistrationof3DUltrasoundImages
UsingModel-basedSegmentation,BabakMatinfar,LyubomirZagorchev1445Poster:ImageRegistrationofVeryLargeImagesvia
GeneticProgramming,SaritChicotay,OmidE.David,NathanS.Netanyahu1455AfternoonBreak
1530InvitedTalk:Representing3DModelswith
DiscriminativeVisualElements,MathieuAubry(INRIA1615EfficientandAutomatedMultimodalSatelliteData
RegistrationThroughMRFsandLinearProgramming,KonstantinosKarantzalos,AristeidisSotiras,NikosParagios
1045Non-rigidPointSetRegistrationwithGlobal-Local
TopologyPreservation,SongGe,GuoliangFan,MengDing11003DSceneEstimationwithPerturbation-Modulated
LightandDistributedSensors,QuanWang,XinchiZhang,KimL.Boyer1115Edge-WeightedCentroidVoronoiTessellationwith
PropagationofConsistencyConstraintfor3DGrainSegmentationinMicroscopicSuperalloyImages,YoujieZhou,LiliJu,YuCao,JarrellWaggoner,YueweiLin,JeffSimmons,SongWangS4:Posters(1130-1230
JointShapeandTextureBasedX-RayCargoImageClassification,JianZhang,LiZhang,ZiranZhao,YaohongLiu,JianpingGu,QiangLi,DuokunZhang
UseofSparseRepresentationforPedestrianDetectioninThermalImages,BinQi,VijayJohn,ZhengLiu,SeiichiMita
APhoton-MappingInformedChan-VeseSegmentationAlgorithmtoEnableMultispectralSensingandPath-Planningin3DVirtualEnvironments,BruceA.Johnson,HairongQi,JasonC.Isaacs
SuperpixelEstimationforHyperspectralImagery,PegahMassoudifar,AnandRangarajan,PaulGaderAutomaticTargetRecognitioninInfraredImagery
UsingDenseHOGFeaturesandRelevanceGroupingofVocabulary,M.N.A.Khan,GuoliangFan,DouglasR.Heisterkamp,LiangjiangYu
Ego-MotionEstimationonRangeImagesusingHigh-OrderPolynomialExpansion,BrianOkorn,JoshHarguess


9


Monday,June23Workshops
1330EffectofPupilDilationandConstrictiononthe
DistributionofBitErrorswithintheIris,InmaculadaTomeo-Reyes,VinodChandran1335OptimizationofIrisCodesforImprovedRecognition,
NitinK.Mahadeo,AndrewP.Papliński,SidRay1340InvitedTalk:ObjectDetectionwithDeepNeural
Network,DumitruErhan(GoogleS2:ExtendedPosterSpotlights(1430-1510
1430ReliablePosteriorProbabilityEstimationforStreaming
FaceRecognition,AbhijitBendale,TerranceBoult1435LearningMinutiaeNeighborhoods:ANewBinary
RepresentationforMatchingFingerprints,AkhilVij,AnoopNamboodiri1440PerformanceImprovementofPhase-Based
CorrespondenceMatchingforPalmprintRecognition,VincentRoux,ShoichiroAoyama,KoichiIto,TakafumiAoki1445ARobustApproachforSingularPointExtractionBased
onComplexPolynomialModel,JinQi,SuxingLiu1450SecureFingerprintMatchingWithGenericLocal
Structures,MatthewMorse,JesseHartloff,ThomasEffland,JimSchuler,JenniferCordaro,SergeyTulyakov,AtriRudra,VenuGovindaraju1455TheValueofMultipleViewpointsinGesture-Based
UserAuthentication,JonathanWu,JanuszKonrad,PrakashIshwar1500Context-AwareActiveAuthenticationUsing
SmartphoneAccelerometerMeasurements,AbenaPrimo,VirV.Phoha,RajeshKumar,AbdulSerwadda1505CanWeUseSecondMinorFingerKnucklePatternsto
IdentifyHumans?,AjayKumar,ZhihuanXu1510AfternoonBreakandPosterSession
1600InvitedTalk:FaceBiometricsunderSpoofingAttacks:
Vulnerabilities,Countermeasures,OpenIssuesandResearchDirections,AbdenourHadid(Univ.ofOulu1650Valedictory,AwardsandClosingRemarks
10
1640Poster:VariationalDeformationMethodforthe
ComputationoftheAverageShapeofOrgans,ShunInagaki,AtsushiImiya1650Poster:AdaptiveRegistrationofVeryLargeImages,
BrianP.Jackson,A.ArdeshirGoshtasby

Biometrics
Organizers:BirBhanuRossBeveridgeAjayKumarLocation:C220-221
Schedule:HalfDay—Afternoon
S1:ExtendedPosterSpotlights(1300-1340
1300HallucinatingtheFullFacefromthePeriocularRegion
viaDimensionallyWeightedK-SVD,FelixJuefei-Xu,DipanK.Pal,MariosSavvides1305Improving3DFaceDetailsbasedonNormalMapof
Hetero-sourceImages,ChangYang,JianshengChen,NanSu,GuangdaSu1310Globality-LocalityPreservingProjectionsforBiometric
DataDimensionalityReduction,ShengHuang,AhmedElgammal,LuwenHuangfu,DanYang,XiaohongZhang1315RobustLow-RankRegularizedRegressionforFace
RecognitionwithOcclusion,JianjunQian,JianYang,FanglongZhang,ZhouchenLin1320NaturalvsArtificialFaceClassificationusingUniform
LocalDirectionalPatternsandWaveletUniformLocalDirectionalPatterns,DarrylD'Souza,RomanV.Yampolskiy1325LandmarkBasedFacialComponentReconstructionfor
RecognitionAcrossPose,Gee-SernHsu,Hsiao-ChiaPeng,Kai-HsiangChang


Monday,June23
DeepLearningforComputerVision
Organizer:GrahamTaylorMarc'AurelioRanzatoHonglakLeeTime:Location:
0830-1700(FullDayExhibitHallC
Tutorials
DenseImageCorrespondencesforComputerVision
Organizer:
Time:Location:
MichaelRubinsteinJaechulKimZhuowenTuCeLiu
0830-1700(FullDayC111-112
Description:Acentralchallengeinvisualreasoningisthatofuntanglingthemanyfactorsofvariationthatexplainanimageorvideo.Photometricandgeometric"nuisance"factorsareintertwinedwiththevariablesofinterest,forexample,objectidentityinrecognitiontasks.Todate,thedominantmethodologyforaddressingthischallengehasbeentoengineerafeatureextractionpipeline,usuallycontainingmultiplestagesofprocessing.Analternativeapproachis"RepresentationLearning":relyingonthedata,insteadoffeatureengineeringtolearnrepresentationsthatareinvarianttonuisancefactors.Techniquesthatlearnmultiplelayersofrepresentation,whicharereferredtoas"DeepLearning",havedemonstratednotonlyimpressivesuccessinrecentbenchmarksandcompetitionsbutapplicabilitytomultipledomains.Thetutorialwillbestructuredintwoparts.Inthemorning,wewillreviewthefoundationsofdeeplearningappliedtovisioninboththesupervisedandunsupervisedsetting.Wewillalsohighlightthemostfrequentlyusedpracticaldevelopmentlibrariesandtools.Intheafternoon,wewillinviteleadingexpertsinthefieldtodiscussthemostrelevantapplicationareas,includingobjectdetection,structuredprediction,large-scaleclassificationandhardwareacceleration,video,multi-modalandmulti-tasklearning,andregressionmethodsforlocalization.
Description:Correspondence,namelyhowpixelsinoneimagecorrespondtopixelsinanotherimage,isafundamentalproblemincomputervision.Althoughcorrespondencehasbeenmostlyusedforanalyzingtransformationsbetweenimagesfromonescene,anewerahasstartedrecentlywhencorrespondencecanbeestablishedacrossdifferentscenes.Inthistutorial,wewillgiveanoverviewofdensecorrespondencealgorithmsforaligningimagesfromdifferentscenes.Wewillsurveyavarietyofrepresentations,includingpixels(SIFTflow,Non-RigidDenseCorrespondence,semanticsegments(layerflowandimagepyramid(deformablespatialpyramid.Thesedensealignmentalgorithmsarepowerfultoolstoanalyzeimagesandvideos.Wecannotonlytransforminformationsuchassemanticlabels,imagedetailsandgeometryfromimagesandvideosinalabeleddataset,butalsoanalyzeanentireimagedatabaseasawholeviainformationpropagation.Recentadvancesonsceneparsing,2Dvideoto3D,annotationpropagation(imagetotext,objectdiscovery,co-segmentation,imagehallucination,andbiomedicalimageanalysisdemonstratethatacross-scenecorrespondencecanbeafundamentalbuildingblockforcomputervision.

11


Monday,June23
Organizer:
Time:Location:
AlexBronsteinMichaelBronsteinIasonasKokkinosGeorgePapandreou0830-1700(FullDayC212

Moreoptimizations(NEON,OpenVX
Tutorials
Moresupportedlanguages(Matlab,Ruby,HaskellAnewandmodularwayofparticipatingtothecoredevelopmentWhileshowcasingtheaforementionedfeatures,wewillfocusbuildingend-to-endvisionpipelinesthroughseveralapplicationwalk-throughs.Code,instructions,andmobileapplicationswillbeavailableonlinebeforethetutorial.
BASIS-14:BASesforImages&Surfaces
Description:BASIS-14willbeafulldaytutorialcoveringthecurrentstate-of-the-artinlinearandnonlinearimageandsurfaceanalysistechniques.Startingwiththefundamentalsoflinearimageprocessing,wewillseehowthemainnotionsofFouriertransformscanbeunderstoodintermsofachangeofbasis,andwillexplorethemultifoldramificationsofthisintuitiontononlinearimageprocessing(sparsecoding,dictionarylearning,exemplarrepresentations,invarianttransformsandsurfaceanalysis(heatdiffusiononsurfaces,spectraldecompositionoftheLaplace-Beltramioperator,surfacedescriptors.
Wewillpresentapplicationstobothclassicalproblems,suchasdenoisinganddeblurring,butalsocuttingedge-computervisionproblemsinvolvingimageclassification,objectdetection,shaperetrieval,andsurfaceregistration.

EmergingTopicsinHumanActivityRecognition
Organizers:MichaelRyooIvanLaptevGregMoriSangminOhTime:Location:
0830-1230(HalfDay—MorningC211

OpenCV3.0:SolvingProblems
Organizer:
Time:Location:
GaryBradski
VadimPisarevskyVincentRabaudGraceVesom0830-1700(FullDayC222
Description:ThethirdmajorreleaseofOpenCVisaimedatbuildingsolidgroundforcomputervisiondevelopment.C++andPythonwillbecoveredhere.OpenCV3.0enables:Morealgorithmstobeintegrated(ofwhichwewillshowcasethelatest
12

Description:Inthepast5years,thefieldofhumanactivityrecognitionhasgrowndramatically,reflectingitsimportanceinmanyhigh-impactsocietalapplicationsincludingsmartsurveillance,web-videosearchandretrieval,quality-of-lifedevicesforelderlypeople,andhuman-computerinterfaces.Giventheinitialsuccessofbag-of-wordsmethodsforactionclassification,thefieldisgraduallymovingtowardsmorestructuredinterpretationofcomplexhumanactivitiesinvolvingmultiplepeopleandobjectsaswellasinteractionsamongtheminvariousrealisticscenarios.Newimportantresearchtopicsandproblemsareappearingasaconsequence,including(imodelingtemporalstructureofactivities,(iilearningrelationsbetweenactionsandobjects/scenes/socialroles,(iiigroupactivityrecognition,and(ivfirst-personactivityrecognition.Theobjectiveofthistutorialistointroduceandoverviewrecentprogressintheseemergingtopics,aswellastodiscuss,motivateandencouragefutureresearchindiversesubfieldsofactivityrecognition.

Monday,June23Tutorials
LearningVisualSemantics:Models,MassiveComputation,&InnovativeApplications
Organizer:
Time:Location:
Shih-FuChangJohnSmithRogerioFerisLiangliangCao

StateoftheArt3DReconstructionTechniques:VeryLargeScale3D
ReconstructionandtheRoleofPriors
Organizers:NoahSnavelyYasutakaFurukawaTime:Location:
0830-1230(HalfDay—MorningC216
Description:Thiscoursewillcoverstate-of-the-art3DreconstructiontechniquesbeyondstandardSfMandMVStechniques,focusingontwokeyareas.Thefirstfocusisinthelargescale3Dreconstruction.AsthecoreStructurefromMotion(SfMandMulti-ViewStereo(MVStechnologiesbecomematureandrobust,moreandmoreinterestsanddemandsaroseforverylarge-scaleSfMandMVSexecutions,primarilyfordigitalmappingapplications.Thesecondfocusisintheuseofstructurepriorsin3Dreconstruction,suchasplanarity,orthogonality,symmetry,andrepetition,whichposechallengestostandardSfMandMVStechniques,butcanyieldrichstructureinformationaboutthesceneforbetter3Dmodeling.
1300-1700(HalfDay—AfternoonC211

Description:Theexplosionofdigitalmultimediadata-includingvisualcontentfromsurveillancecameras,mobilephones,personalphotocollections,newsfootage,ormedicalimages–iscreatingsignificantopportunitiesforautomatedvisualanalysis.However,themostinterestingcontentinmultimediafilesisoftenunconstrainedandcomplexinnature,reflectingadiversityofhumanbehaviors,scenes,activities,andevents,whichposesseriouschallengesforcomputervisionapproaches.Inthistutorial,wewillpresentthestate-of-the-artonlarge-scalevisualsemanticmodeling,coveringmethodsforobtainingintuitivemid-levelsemanticfeaturerepresentations,whilepresentinginnovativeapplications.Theorganizerswillsharetheirexperienceinachievingtopperformanceonseveralrecentcompetitions,includingTRECVID,ImageNet,andImageCLEF,anddevelopinglarge-scaledataandtoolresources.
13


Monday,June23
VideoSegmentation
Organizer:JasonCorsoMatthiasGrundmanIrfanEssaTime:Location:
1300-1700(HalfDay—AfternoonC213
Tutorials
Large-ScaleVisualPlaceRecognitionandImage-BasedLocalization
Organizer:TorstenSattlerAkihikoToriiTime:Location:
1300-1700(HalfDay—AfternoonC216
Description:Inrecentyears,segmentationhasemergedasaplausiblefirststepinearlyvideoprocessingofunconstrainedvideos,withoutneedingtomakeanassumptionofastaticbackgroundasearliermethodshave.Videosegmentationandover-segmentation,ormorecommonlysupervoxelextrac-tion,isacomplementaryearlyvideoprocessingsteptothemoretraditionalfeatureextraction,suchasSTIPandtrajecto-ries,anditextendsthelonghistoryofimagesegmentationmethods.Thistutorialwillsurveyandpresenttheimportantmodelsandalgorithmsforvideosegmentation.Wewillcoverdirectextensionsofimagesegmentationmethodsthroughvideo-specificspatiotemporalandstreamingmethods.Inadditiontocoremethodologicalelements,thetutorialwillalsocoverbenchmarkandevaluationofvideosegmentationaswellasapplicationsofvideosegmentation.ParticipantswillbeintroducedtothedetailsofthesemethodsnotonlythroughtraditionalslidepresentationsbutalsoexampleimplementationsthroughtheLIBSVXlibrary.

Description:Thetutorialconsistsoftwopartscoveringthegeneralproblemsofvisualplacerecognitionandimage-basedlocalization.Thefirstpartisaboutvisualplacerecognitionandconsidersanapplicationscenarioinwhichthesceneisrepresentedbyasetofgeo-taggedimages.Theaimofvisualplacerecognitionistoapproximatethepositionoftheviewerbyidentifyingtheplacevisibleinthequeryimageusing(imageretrievalmethods.Wediscussseveralimprovementstothestandardretrievalpipelinethatdetectandremoveconfusingfeatures,exploittheknownspatialrelationsbetweentheimages,incorporatepriorsontheviewer’sposition,andenableplacerecognitionsystemstohandletherepetitivestructuresprevalentinurbanenvironments.Thesecondpartofthetutorialisaboutimage-basedlocalizationandconsidersthemorespecifictaskofpreciselyestimatingtheposeofthequeryimagerelativetoa3Dmodelofthescene.Assumingthatthis3DmodelwasreconstructedusingStructure-from-Motion,wecanfindcorrespondencesbetween2Dfeaturesinthequeryimageand3Dpointsinthemodelusingdescriptormatching.Wefirstintroducethestandarddatastructuresfordescriptormatchingaswellasdifferentapproachestoestimatethecameraposefromthe2D-3Dmatches.Wethendetailtheprioritizedmatchingschemesthatenablestate-of-the-artlocalizationsystemstoefficientlyhandle3Dmodelsconsistingofmillionsof3Dpoints.Wealsodiscusshowtoexploitexistingvisibilityinformationbetween3Dpointsinthemodelandthedatabaseimagesandhowtoreducethememoryrequirementsbyusingonlyasubsetofall3Dpointswithoutlossoflocalizationperformance.Throughoutthetutorial,weprovidelinkstopubliclyavailablesourcecodeforthediscussedapproachesaswellaspubliclyavailabledatasets.
14


Tuesday,June24(MorningProgram
0830–1000Oral1B:Segmentation&Grouping
(BattelleGrandNorth
PosterIDsforthissession:O-1B-#where#isthepaper#.Chairs:PiotrDollar(MicrosoftResearchTalArbel(McGillUniv.
1.
Tuesday,June24
0700–1700Registration(ExhibitHallCLobby0730–0830Breakfast(ExhibitHallC
0820–0830WelcomebytheGeneralChairs
(BattelleGrand
2.
3.4.
0830-1200AMVideoSpotlights(C213-215
0830–1000Oral1A:Matching&Reconstruction
(BattelleGrandSouth
PosterIDsforthissession:O-1A-#where#isthepaper#.Chairs:JanaKosecka(GeorgeMasonUniv.AntonisArgyros(Univ.ofCrete
1.
5.6.
Format(13min.forpresentation+2min.forquestions
SpectralGraphReductionforEfficientImageand
StreamingVideoSegmentation,FabioGalasso,MargretKeuper,ThomasBrox,BerntSchiele
WeaklySupervisedMulticlassVideoSegmentation,XiaoLiu,DachengTao,MingliSong,YingRuan,ChunChen,JiajunBu
VideoMotionSegmentationUsingNewAdaptiveManifoldDenoisingModel,DijunLuo,HengHuang
Cut,Glue&Cut:AFast,ApproximateSolverforMulticutPartitioning,ThorstenBeier,ThorbenKroeger,JörgH.Kappes,UllrichKöthe,FredA.Hamprecht
NeuralDecisionForestsforSemanticImageLabelling,SamuelRotaBulò,PeterKontschieder
PullingThingsoutofPerspective,ĽuborLadický,JianboShi,MarcPollefeys
1000–1030Break(GrandBallroomPrefunction1000–1200Exhibits(GrandBallrooms1-3

15
2.3.4.5.
6.
Format(13min.forpresentation+2min.forquestionsFastandAccurateImageMatchingwithCascadeHashingfor3DReconstruction,JianCheng,CongLeng,JiaxiangWu,HainanCui,HanqingLu
PredictingMatchability,WilfriedHartmann,MichalHavlena,KonradSchindler
TrinocularGeometryRevisited,JeanPonce,MartialHebertCriticalConfigurationsForRadialDistortionSelf-Calibration,ChangchangWu
SolversforRelativePosewithaSingleUnknownRadialDistortionMinimal,YubinKuang,JanErikSolem,FredrikKahl,KalleÅström
ReconstructingPASCALVOC,SaraVicente,JoãoCarreira,LourdesAgapito,JorgeBatista
MicrosoftGoogleXeroxAmazonNVIDIAA9Face++MetaioIntelCuralateIBMOrbeusOMRONQualcommItseez,Inc•••••••••••••••CogniVueElsevierMathWorksPointGreynowpublishers
CRCPress/Taylor&FrancisMorgan&ClaypoolSpringerKAUSTApple,Inc3dMD
OccamVisionGroupEyerisSpotscale
SamsungMPILab


Tuesday,June24(Morning
1000–1200Demos(C110-115
Program
11.FromCategoriestoIndividualsinRealTime—AUnified
TurningMobilePhonesinto3DScanners,PetriTanskanen,KalinKolev,AmaelDelaunoy,MarcPollefeys(ETHZurichVirtualMakeup,SifeiLiu,JimeiYang,ZheHu(UCMercedLearningtobeaDepthCamera,SeanRyanFanello,CemKeskin,ShahramIzadi,PushmeetKohli,DavidKim,DavidSweeney,AntonioCriminisi,JamieShotton,SingBingKang,TimPaek(MicrosoftResearch
FilterForestforLearningData-DependentFilters,SeanRyanFanello,CemKeskin,PushmeetKohli,ShahramIzadi,JamieShotton,AntonioCriminisi,TimPaek(MicrosoftResearch
1000–1200Poster1A:Recognition,
Segmentation,Stereo&SFM(GrandBallrooms1-3
PosterIDsforthissession:P-1A-#where#isthepaper#.1.EventDetectionusingMulti-LevelRelevanceLabelsandMultipleFeatures,ZhongwenXu,IvorW.Tsang,YiYang,ZhigangMa,AlexanderG.Hauptmann
2.Full-AngleQuaternionsforRobustlyMatchingVectorsof3DRotations,StephanLiwicki,Minh-TriPham,StefanosZafeiriou,MajaPantic,BjörnStenger
3.Humanvs.ComputerinSceneandObjectRecognition,AliBorji,LaurentItti
4.Semi-supervisedSpectralClusteringforImageSet
Classification,ArifMahmood,AjmalMian,RobynOwens5.LookattheDriver,LookattheRoad:NoDistraction!NoAccident!,MahdiRezaei,ReinhardKlette
6.MeasuringDistanceBetweenUnorderedSetsofDifferentSizes,AndrewGardner,JinkoKanno,ChristianA.Duncan,RastkoSelmic
7.LearningMid-levelFiltersforPersonRe-identification,RuiZhao,WanliOuyang,XiaogangWang
8.DeepReID:DeepFilterPairingNeuralNetworkforPersonRe-Identification,WeiLi,RuiZhao,TongXiao,XiaogangWang
9.LacunarityAnalysisonImagePatternsforTexture
Classification,YuhuiQuan,YongXu,YupingSun,YuLuo10.Segmentation-awareDeformablePartModels,EduardTrulls,StavrosTsogkas,IasonasKokkinos,AlbertoSanfeliu,FrancescMoreno-Noguer
16

BoostingApproach,DavidHall,PietroPerona
12.NMF-KNN:ImageAnnotationusingWeightedMulti-viewNon-negativeMatrixFactorization,MahdiM.Kalayeh,HaroonIdrees,MubarakShah
13.Fine-GrainedVisualComparisonswithLocalLearning,AronYu,KristenGrauman
14.InferringAnalogousAttributes,Chao-YehChen,KristenGrauman
15.BeyondComparingImagePairs:SetwiseActiveLearningforRelativeAttributes,LucyLiang,KristenGrauman16.VisualPersuasion:InferringCommunicativeIntentsofImages,JungseockJoo,WeixinLi,FrancisF.Steen,Song-ChunZhu
17.HistogramsofPatternSetsforImageClassificationandObjectRecognition,WinnVoravuthikunchai,BrunoCrémilleux,FrédéricJurie
18.IncorporatingSceneContextandObjectLayoutinto
AppearanceModeling,HamidIzadinia,FereshtehSadeghi,AliFarhadi
19.Co-SegmentationofTextured3DShapeswithSparseAnnotations,MehmetErsinYumer,WonChun,AmeeshMakadia
20.HowtoEvaluateForegroundMaps?,RanMargolin,LihiZelnik-Manor,AyelletTal
21.MILCut:ASweepingLineMultipleInstanceLearning
ParadigmforInteractiveImageSegmentation,JiajunWu,YibiaoZhao,Jun-YanZhu,SiweiLuo,ZhuowenTu
22.SCAMS:SimultaneousClusteringandModelSelection,ZhuwenLi,Loong-FahCheong,StevenZhiyingZhou23.TheShape-TimeRandomFieldforSemanticVideo
Labeling,AndrewKae,BenjaminMarlin,ErikLearned-Miller24.TheSecretsofSalientObjectSegmentation,YinLi,XiaodiHou,ChristofKoch,JamesM.Rehg,AlanL.Yuille
25.Non-rigidSegmentationusingSparseLowDimensionalManifoldsandDeepBeliefNetworks,JacintoC.Nascimento,GustavoCarneiro
26.AnExemplar-basedCRFforMulti-instanceObjectSegmentation,XumingHe,StephenGould
27.ObjectPartitioningusingLocalConvexity,SimonChristophStein,MarkusSchoeler,JeremiePapon,FlorentinWörgötter

Tuesday,June24(Morning
28.BayesianActiveContourswithAffine-Invariant,Elastic
Program
44.PartialSymmetryinPolynomialSystemsandits
ShapePrior,DarshanBryner,AnujSrivastava29.Max-MarginBoltzmannMachinesforObject
Segmentation,JimeiYang,SimonSafar,Ming-HsuanYang30.MultiscaleCombinatorialGrouping,PabloArbeláez,JordiPont-Tuset,JonathanT.Barron,FerranMarques,JitendraMalik
31.RIGOR:ReusingInferenceinGraphCutsforGeneratingObjectRegions,AhmadHumayun,FuxinLi,JamesM.Rehg32.EfficientHierarchicalGraph-BasedSegmentationofRGBDVideos,StevenHickson,StanBirchfield,IrfanEssa,HenrikChristensen
33.PointMatchinginthePresenceofOutliersinBothPointSets:AConcaveOptimizationApproach,WeiLian,LeiZhang
34.MultipleStructured-InstanceLearningforSemanticSegmentationwithUncertainTrainingData,Feng-JuChang,Yen-YuLin,Kuang-JuiHsu
35.JointMotionSegmentationandBackgroundEstimationinDynamicScenes,AdeelMumtaz,WeichenZhang,AntoniB.Chan
36.SeamSeg:VideoObjectSegmentationusingPatchSeams,S.AvinashRamakanth,R.VenkateshBabu
37.LaplacianCoordinatesforSeededImageSegmentation,WallaceCasaca,LuisGustavoNonato,GabrielTaubin38.Error-tolerantScribblesBasedInteractiveImageSegmentation,JunjieBai,XiaodongWu
39.IterativeMultilevelMRFLeveragingContextandVoxelInformationforBrainTumourSegmentationinMRI,NageshSubbanna,DoinaPrecup,TalArbel
40.LargeScaleMulti-viewStereopsisEvaluation,RasmusJensen,AndersDahl,GeorgeVogiatzis,EnginTola,HenrikAanæs
41.Timing-BasedLocalDescriptorforDynamicSurfaces,TonyTung,TakashiMatsuyama
42.AMinimalSolutiontotheGeneralizedPose-and-ScaleProblem,JonathanVentura,ClemensArth,GerhardReitmayr,DieterSchmalstieg
43.AGeneralandSimpleMethodforCameraPoseandFocalLengthDetermination,YinqiangZheng,ShigekiSugimoto,ImariSato,MasatoshiOkutomi
ApplicationsinComputerVision,YubinKuang,YinqiangZheng,KalleÅström
45.EfficientComputationofRelativePoseforMulti-CameraSystems,LaurentKneip,HongdongLi
46.SimultaneousLocalizationandCalibration:Self-CalibrationofConsumerDepthCameras,Qian-YiZhou,VladlenKoltun
47.MinimalSceneDescriptionsfromStructurefromMotionModels,SongCao,NoahSnavely
48.Fast,ApproximatePiecewise-PlanarModelingBasedonSparseStructure-from-MotionandSuperpixels,AndrásBódis-Szomorú,HaykoRiemenschneider,LucVanGool49.OnProjectiveReconstructionInArbitraryDimensions,BehroozNasihatkon,RichardHartley,JochenTrumpf50.StereounderSequentialOptimalSampling:AStatisticalAnalysisFrameworkforSearchSpaceReduction,YilinWang,KeWang,EnriqueDunn,Jan-MichaelFrahm
51.EfficientPruningLMIConditionsforBranch-and-PruneRankandChirality-ConstrainedEstimationoftheDualAbsoluteQuadric,AdlaneHabed,DandaPaniPaudel,CédricDemonceaux,DavidFofi
52.VeryFastSolutiontothePnPProblemwithAlgebraicOutlierRejection,LuisFerraz,XavierBinefa,FrancescMoreno-Noguer
53.FindingVanishingPointsviaPointAlignmentsinImagePrimalandDualDomains,JoséLezama,RafaelGromponevonGioi,GregoryRandall,Jean-MichelMorel
54.DiscriminativeFeature-to-PointMatchinginImage-BasedLocalization,MichaelDonoser,DieterSchmalstieg
55.Two-ViewCameraHousingParametersCalibrationforMulti-LayerFlatRefractiveInterface,XidaChen,Yee-HongYang
56.AccurateLocalizationandPoseEstimationforLarge3DModels,LinusSvärm,OlofEnqvist,MagnusOskarsson,FredrikKahl
57.RelativePoseEstimationforaMulti-CameraSystemwithKnownVerticalDirection,GimHeeLee,MarcPollefeys,FriedrichFraundorfer
1200–1330Lunch(ExhibitHallC
17


Tuesday,June24(MorningProgram
CewuLu(HongKonkUniv.ofScience&TechnologyPingLuo(TheChineseUniv.ofHongKongMohammadNorouzi(Univ.ofTorontoIasonOikonomidis(Univ.ofCreteMattO'Toole(Univ.ofToronto
JeremiePapon(Georg-August-UniversitätGöttingenHyunSooPark(CarnegieMellonUniv.BryanPoling(Univ.ofMinnesota
VittalPremachandran(NationalUniv.ofSingaporeLiuShuaicheng(NationalUniv.ofSingaporeEranSwears(RensselaerPolytechnicInst.DanhangTang(ImperialCollegeLondonEduardTrulls(CSIC-UPC
XiaoyangWang(RensselaerPolytechnicInst.ZhaowenWang(Univ.ofIllinois,Urbana-ChampaignChih-YuanYang(Univ.ofCalifornia,MercedJimeiYang(Univ.ofCalifornia,MercedJinweiYe(Univ.ofDelawareQuanshiZhang(Univ.ofTokyo
YingyingZhu(Univ.ofQueensland&CSIROLiZhuwen(NationalUniv.ofSingapore
1200–1330DoctoralConsortium(C210-212
(byinvitationonly
Supportedby:

GuangChen(Univ.ofMissouriYuChen(NorthwesternUniv.
BaoChenglong(NationalUniv.ofSingapore
R.GokberkCinbis(INRIARhone-Alpes&Univ.ofGrenobleJifengDai(TsinghuaUniv.
JianDong(NationalUniv.ofSingaporeKunDuan(IndianaUniv.AlexanderFix(CornellUniv.
VictorM.Fragoso(Univ.ofCalifornia,SantaBarbaraEfstratiosGavves(Univ.ofAmsterdamYunchaoGong(Univ.ofNorthCarolina
AbnerGuzmán-Rivera(Univ.ofIllinois,Urbana-ChampaignHanHu(TsinghuaUniv.
ZheHu(Univ.ofCalifornia,MercedZhiwuHuang(ChineseAcademyofSciencesSatoshiIkehata(Univ.ofTokyoCatalinIonescu(Univ.ofBonn
SadeepJayasumana(AustralianNationalUniv.AndrewKae(Univ.ofMassachusettsAmherstLeKang(Univ.ofMarylandVahidKazemi(KTHGimHeeLee(ETHZürich
WenLi(NanyangTechnologicalUniv.StephanLiwicki(ImperialCollegeLondon
18



Tuesday,June24(Afternoon
1330-1830PMVideoSpotlights(C213-215
1330–1445Oral1C:StatisticalMethods&
LearningI(BattelleGrandSouth
PosterIDsforthissession:O-1C-#where#isthepaper#.Chairs:RaquelUrtasun(Univ.ofTorontoAndreaVedaldi(Univ.ofOxford
1.2.3.
Program
1445–1515Break(BattelleGrandPrefunction1515–1630Special1:Awards&PlenarySession
(BattelleGrand
Chairs:AleixMartinez(OhioStateUniv.CorneliaFermüller(Univ.ofMarylandAwardsCeremony:ProgramChairs
PlenaryTalk:NeuralMechanismsforFaceProcessing,DorisTsao(CaliforniaInst.ofTechnology
Abstract:Howthebraindistillsarepresentationof
meaningfulobjectsfromretinalinputisoneofthecentralchallengesofsystemsneuroscience.Functionalimagingexperimentsinthemacaquerevealthatoneecologicallyimportantclassofobjects,faces,isrepresentedbyasystemofsixdiscrete,stronglyinterconnectedregions.Electrophysiologicalrecordingsshowthatthese'facepatches'haveuniquefunctionalprofiles.Bystudyingthedistinctvisualrepresentationsmaintainedinthesesixfacepatches,thesequenceofinformationflowbetweenthem,andtheroleeachplaysinfaceperception,wearegainingnewinsightsintohierarchicalinformationprocessinginthebrain.
4.5.
Format(13min.forpresentation+2min.forquestionsOptimalDecisionsfromProbabilisticModels:TheIntersection-over-UnionCase,SebastianNowozinCovarianceTreesfor2Dand3DProcessing,ThierryGuillemot,AndrésAlmansa,TamyBoubekeur
HierarchicalSubqueryEvaluationforActiveLearningonaGraph,OisinMacAodha,NeillD.F.Campbell,JanKautz,GabrielJ.Brostow
AnytimeRecognitionofObjectsandScenes,SergeyKarayev,MarioFritz,TrevorDarrell
RichFeatureHierarchiesforAccurateObjectDetectionandSemanticSegmentation,RossGirshick,JeffDonahue,TrevorDarrell,JitendraMalik
1330–1445Oral1D:ActionRecognition
(BattelleGrandNorth
PosterIDsforthissession:O-1D-#where#isthepaper#.Chairs:SinisaTodorovic(OregonStateUniv.SudeepSarkar(Univ.ofSouthFlorida
1.
1630–1830Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
1630–1830Demos(C110-115
ObjectPartioningusingLocalConvexity,MarkusSchoeler,
JeremiePapon(Univ.ofGöttingen
AnalysisbySynthesis:3DObjectRecognitionbyObjectReconstruction,MohsenHejrati,DevaRamanan(UCIrvineEstimatingImageDepthUsingShapeCollections,HaoSu,QixingHuang,NiloyMitra,YangyanLi,LeonidasGuibas(Stanford,Univ.CollegeLondon
VisipediaBackend:CollaborativeToolsforImageDatasetCreationandManagement,GrantVanHorn,Steve
Branson,CatherineWah,PietroPerona,SergeBelongie(UCSanDiego,Caltech,CornellTech
2.3.4.5.
Format(13min.forpresentation+2min.forquestionsHumanActionRecognitionbyRepresenting3DSkeletonsasPointsinaLieGroup,RavitejaVemulapalli,FelipeArrate,RamaChellappa
Multi-ViewSuperVectorforActionRecognition,ZhuoweiCai,LiminWang,XiaojiangPeng,YuQiao
UnsupervisedSpectralDualAssignmentClusteringofHumanActionsinContext,SimonJones,LingShaoParsingVideosofActionswithSegmentalGrammars,HamedPirsiavash,DevaRamanan
Rate-InvariantAnalysisofTrajectoriesonRiemannianManifoldswithApplicationinVisualSpeechRecognition,JingyongSu,AnujSrivastava,FillipeD.M.deSouza,SudeepSarkar
19



Tuesday,June24(Afternoon
1630–1830Poster1B:3DVision,Action
Recognition,Recognition,StatisticalMethods&Learning(GrandBallrooms1-3
PosterIDsforthissession:P-1B-#where#isthepaper#.1.PiecewisePlanarandCompactFloorplanReconstructionfromImages,RicardoCabral,YasutakaFurukawa
2.Data-drivenFlowerPetalModelingwithBotanyPriors,ChenxiZhang,MaoYe,BoFu,RuigangYang
3.User-SpecificHandModelingfromMonocularDepthSequences,JonathanTaylor,RichardStebbing,VarunRamakrishna,CemKeskin,JamieShotton,ShahramIzadi,AaronHertzmann,AndrewFitzgibbon
4.ClassSpecific3DObjectShapePriorsUsingSurface
Normals,ChristianHäne,NikolaySavinov,MarcPollefeys5.Frequency-Based3DReconstructionofTransparentandSpecularObjects,DingLiu,XidaChen,Yee-HongYang6.HumanBodyShapeEstimationUsingaMulti-ResolutionManifoldForest,FrankPerbet,SamJohnson,Minh-TriPham,BjörnStenger
7.QualityDynamicHumanBodyModelingUsingaSingleLow-costDepthCamera,QingZhang,BoFu,MaoYe,RuigangYang
8.Single-View3DSceneParsingbyAttributedGrammar,XiaobaiLiu,YibiaoZhao,Song-ChunZhu
9.SeparationofLineDrawingsBasedonSplitFacesfor3DObjectReconstruction,ChangqingZou,HengYang,JianzhuangLiu
10.When3DReconstructionMeetsUbiquitousRGB-D
Images,QuanshiZhang,XuanSong,XiaoweiShao,HuijingZhao,RyosukeShibasaki
11.StableTemplate-BasedIsometric3DReconstructioninAllImagingConditionsbyLinearLeast-Squares,AjadChhatkuli,DanielPizarro,AdrienBartoli
12.Discrete-ContinuousDepthEstimationfromaSingleImage,MiaomiaoLiu,MathieuSalzmann,XumingHe
13.LeveragingHierarchicalParametricNetworksforSkeletalJointsBasedActionSegmentationandRecognition,DiWu,LingShao
14.SeeingWhatYou'reTold:Sentence-GuidedActivityRecognitionInVideo,NarayanaswamySiddharth,AndreiBarbu,JeffreyMarkSiskind
20

Program
15.ActionLocalizationwithTubeletsfromMotion,MihirJain,
JanvanGemert,HervéJégou,PatrickBouthemy,CeesG.M.Snoek
16.ActionnessRankingwithLatticeConditionalOrdinal
RandomFields,WeiChen,CaimingXiong,RanXu,JasonJ.Corso
17.MultipleGranularityAnalysisforFine-grainedAction
Detection,BingbingNi,VigneshR.Paramathayalan,PierreMoulin
18.HumanActionRecognitionAcrossDatasetsby
Foreground-weightedHistogramDecomposition,WaqasSultani,ImranSaleemi
19.Range-SampleDepthFeatureforActionRecognition,CewuLu,JiayaJia,Chi-KeungTang
20.TheLanguageofActions:RecoveringtheSyntaxandSemanticsofGoal-DirectedHumanActivities,HildeKuehne,AliArslan,ThomasSerre
21.ComplexActivityRecognitionusingGrangerConstrainedDBN(GCDBNinSportsandSurveillanceVideo,EranSwears,AnthonyHoogs,QiangJi,KimBoyer
22.IncrementalActivityModelingandRecognitioninStreamingVideos,MahmudulHasan,AmitK.Roy-Chowdhury
23.SuperNormalVectorforActivityRecognitionUsingDepthSequences,XiaodongYang,YingLiTian
24.DiscriminativeHierarchicalModelingofSpatio-TemporallyComposableHumanActivities,IvanLillo,AlvaroSoto,JuanCarlosNiebles
25.AMultigraphRepresentationforImproved
Unsupervised/Semi-supervisedLearningofHumanActions,SimonJones,LingShao
26.StoryGraphs:VisualizingCharacterInteractionsasaTimeline,MakarandTapaswi,MartinBäuml,RainerStiefelhagen
27.LearningReceptiveFieldsforPoolingfromTensorsofFeatureResponse,CanXu,NunoVasconcelos
28.TowardsUnifiedHumanParsingandPoseEstimation,JianDong,QiangChen,XiaohuiShen,JianchaoYang,ShuichengYan
29.AsktheImage:SupervisedPoolingtoPreserveFeatureLocality,SeanRyanFanello,NicolettaNoceti,CarloCiliberto,GiorgioMetta,FrancescaOdone

Tuesday,June24(Afternoon
30.SimilarityComparisonsforInteractiveFine-Grained
Program
47.SimultaneousTwinKernelLearningusingPolynomial
Categorization,CatherineWah,GrantVanHorn,SteveBranson,SubhransuMaji,PietroPerona,SergeBelongie31.ContinuousManifoldBasedAdaptationforEvolvingVisualDomains,JudyHoffman,TrevorDarrell,KateSaenko
32.TalkingHeads:DetectingHumansandRecognizingTheirInteractions,MinhHoai,AndrewZisserman
33.SalientRegionDetectionviaHigh-DimensionalColorTransform,JiwhanKim,DongyoonHan,Yu-WingTai,JunmoKim
34.TheRoleofContextforObjectDetectionandSemanticSegmentationintheWild,RoozbehMottaghi,XianjieChen,XiaobaiLiu,Nam-GyuCho,Seong-WhanLee,SanjaFidler,RaquelUrtasun,AlanYuille
35.SwitchableDeepNetworkforPedestrianDetection,PingLuo,YonglongTian,XiaogangWang,XiaoouTang36.CompactRepresentationforImageClassification:ToChooseortoCompress?,YuZhang,JianxinWu,JianfeiCai37.CapturingLong-tailDistributionsofObjectSubcategories,XiangxinZhu,DragomirAnguelov,DevaRamanan38.AccurateObjectDetectionwithJointClassification-RegressionRandomForests,SamuelSchulter,ChristianLeistner,PaulWohlhart,PeterM.Roth,HorstBischof39.AdditiveQuantizationforExtremeVectorCompression,ArtemBabenko,VictorLempitsky
40.ProductSparseCoding,TiezhengGe,KaimingHe,JianSun41.InformedHaar-likeFeaturesImprovePedestrian
Detection,ShanshanZhang,ChristianBauckhage,ArminB.Cremers
42.ImageReconstructionfromBag-of-Visual-Words,HiroharuKato,TatsuyaHarada
43.BetaProcessMultipleKernelLearning,BingbingNi,TengLi,PierreMoulin
44.RandomLaplaceFeatureMapsforSemigroupKernelsonHistograms,JiyanYang,VikasSindhwani,QuanfuFan,HaimAvron,MichaelW.Mahoney
45.Hash-SVM:ScalableKernelMachinesforLarge-ScaleVisualClassification,YadongMu,GangHua,WeiFan,Shih-FuChang
46.TransitiveDistanceClusteringwithK-MeansDuality,ZhidingYu,ChunjingXu,DeyuMeng,ZhuoHui,FanyiXiao,WenboLiu,JianzhuangLiu
21

TransformationsforStructuredPrediction,ChetanTonde,AhmedElgammal
48.BregmanDivergencesforInfiniteDimensionalCovarianceMatrices,MehrtashHarandi,MathieuSalzmann,FatihPorikli
49.OptimizingAveragePrecisionusingWeaklySupervisedData,AseemBehl,C.V.Jawahar,M.PawanKumar
50.SubspaceClusteringforSequentialData,StephenTierney,JunbinGao,YiGuo
51.PredictingMultipleAttributesviaRelativeMulti-taskLearning,LinChen,QiangZhang,BaoxinLi
52.LearningInhomogeneousFRAMEModelsforObjectPatterns,JianwenXie,WenzeHu,Song-ChunZhu,YingNianWu
53.EmpiricalMinimumBayesRiskPrediction:HowtoExtractanExtraFew%PerformancefromVisionModelswithJustThreeMoreParameters,VittalPremachandran,DanielTarlow,DhruvBatra
54.FantopeRegularizationinMetricLearning,MarcT.Law,NicolasThome,MatthieuCord
55.Kernel-PCAAnalysisofSurfaceNormalsforShape-from-Shading,PatrickSnape,StefanosZafeiriou
56.MergingSVMswithLinearDiscriminantAnalysis:ACombinedModel,SymeonNikitidis,StefanosZafeiriou,MajaPantic
57.StableLearninginCodingSpaceforMulti-ClassDecodingandItsExtensionforMulti-ClassHypothesisTransferLearning,BangZhang,YiWang,YangWang,FangChen58.FindingtheSubspaceMeanorMediantoFitYourNeed,TimMarrinan,J.RossBeveridge,BruceDraper,MichaelKirby,ChrisPeterson

1830–2030Reception(BattelleGrand

Wednesday,June25(MorningProgram
0830–1000Oral2B:DiscreteOptimization
(BattelleGrandNorth
PosterIDsforthissession:O-2B-#where#isthepaper#.Chairs:OlgaVeksler(Univ.ofWesternOntarioHiroshiIshikawa(WasedaUniv.
1.
Wednesday,June25
0700–1700Registration(ExhibitHallCLobby0730–0830Breakfast(ExhibitHallC
0830-1200AMVideoSpotlights(C213-215
0830–1000Oral2A:Motion&Tracking
(BattelleGrandSouth
PosterIDsforthissession:O-2A-#where#isthepaper#.Chairs:SimonLucey(CSIROMing-HsuanYang(UCMerced
Format(13min.forpresentation+2min.forquestions
1.AdaptiveColorAttributesforReal-TimeVisualTracking,
2.3.
4.5.
6.
Format(13min.forpresentation+2min.forquestions
APrimal-DualAlgorithmforHigher-OrderMultilabelMarkovRandomFields,AlexanderFix,ChenWang,RaminZabih
EnergyBasedMulti-modelFitting&Matchingfor3DReconstruction,HossamIsack,YuriBoykov
SubmodularizationforBinaryPairwiseEnergies,LenaGorelick,YuriBoykov,OlgaVeksler,IsmailBenAyed,AndrewDelong
MaximumPersistencyinEnergyMinimization,AlexanderShekhovtsov
PartialOptimalitybyPruningforMAP-inferencewithGeneralGraphicalModels,PaulSwoboda,BogdanSavchynskyy,JörgH.Kappes,ChristophSchnörrSceneLabelingUsingBeamSearchUnderMutexConstraints,AnirbanRoy,SinisaTodorovic
2.3.4.
5.6.
MartinDanelljan,FahadShahbazKhan,MichaelFelsberg,JoostvandeWeijer
LocalLayeringforJointMotionEstimationandOcclusionDetection,DeqingSun,CeLiu,HanspeterPfister
RealtimeandRobustHandTrackingfromDepth,ChenQian,XiaoSun,YichenWei,XiaoouTang,JianSun
Multi-OutputLearningforCameraRelocalization,AbnerGuzman-Rivera,PushmeetKohli,BenGlocker,JamieShotton,TobySharp,AndrewFitzgibbon,ShahramIzadiMAPVisibilityEstimationforLarge-ScaleDynamic3DReconstruction,HanbyulJoo,HyunSooPark,YaserSheikhMulti-ObjectTrackingviaConstrainedSequentialLabeling,ShengChen,AlanFern,SinisaTodorovic
1000–1030Break(GrandBallroomPrefunction1000–1200Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
1000–1200Demos(C110-115
AutomaticFaçadeParsingandLOD3ModelGenerationfrom3DPointClouds,WilliamNguatem,MartinDrauschke,HelmutMayer(BundeswehrUniv.Munich
Story-basedVideoRetrievalinTVseriesusingPlotSynopses,MakarandTapaswi,MartinBäuml,RainerStiefelhagen(KarlsruheInst.ofTechnologyAuthenticationUsingSketcheswithBiometeric
Information,BenjaminS.Riggan,WesleyE.Snyder,CliffWang(NCStateUniv.,USArmyResearchOfficeTrackingMultipleInteractingTargetsinaCameraNetwork,ShuZhang,AmitK.Roy-Chowdhury(UCRiverside
22



Wednesday,June25(Morning
1000–1200Poster2A:Motion&Tracking,
Optimization,StatisticalMethods&Learning,Stereo&SFM(GrandBallrooms1-3
PosterIDsforthissession:P-2A-#where#isthepaper#.1.PersistentTrackingforWideAreaAerialSurveillance,JanProkaj,GérardMedioni
2.Multi-CueVisualTrackingUsingRobustFeature-LevelFusionBasedonJointSparseRepresentation,XiangyuanLan,AndyJ.Ma,PongC.Yuen
3.Multi-ForestTracker:AChameleoninTracking,DavidJosephTan,SlobodanIlic
4.RigidMotionSegmentationusingRandomizedVoting,HeechulJung,JeongwooJu,JunmoKim
5.RobustOnlineMulti-ObjectTrackingbasedonTrackletConfidenceandOnlineDiscriminativeAppearanceLearning,Seung-HwanBae,Kuk-JinYoon
6.Pyramid-basedVisualTrackingUsingSparsity
RepresentedMeanTransform,ZheZhang,KinHongWong7.TrackletAssociationwithOnlineTarget-SpecificMetricLearning,BingWang,GangWang,KapLukChan,LiWang8.AnOnlineLearnedElementaryGroupingModelforMulti-targetTracking,XiaojingChen,ZhenQin,LeAn,BirBhanu9.Diversity-EnhancedCondensationAlgorithmandItsApplicationforRobustandAccurateEndoscopeThree-DimensionalMotionTracking,XiongbiaoLuo,YingWan,XiangjianHe,JieYang,KensakuMori
10.PartialOcclusionHandlingforVisualTrackingviaRobustPartMatching,TianzhuZhang,KuiJia,ChangshengXu,YiMa,NarendraAhuja
11.SpeedingUpTrackingbyIgnoringFeatures,LuZhang,HamdiDibeklioğlu,LaurensvanderMaaten
12.SubspaceTrackingunderDynamicDimensionalityforOnlineBackgroundSubtraction,MatthewBerger,LeeM.Seversky
13.MultipleTargetTrackingBasedonUndirectedHierarchicalRelationHypergraph,LongyinWen,WenboLi,JunjieYan,ZhenLei,DongYi,StanZ.Li
14.Bi-labelPropagationforGenericMultipleObjectTracking,WenhanLuo,Tae-KyunKim,BjörnStenger,XiaoweiZhao,RobertoCipolla
23

Program
15.AProbabilisticFrameworkforMultitargetTrackingwith
MutualOcclusions,MenglongYang,YiguangLiu,LongyinWen,ZhishengYou,StanZ.Li
16.OcclusionGeodesicsforOnlineMulti-ObjectTracking,HorstPossegger,ThomasMauthner,PeterM.Roth,HorstBischof
17.EfficientNonlinearMarkovModelsforHumanMotion,AndreasM.Lehrmann,PeterV.Gehler,SebastianNowozin18.ACompositionalModelforLow-DimensionalImageSetRepresentation,HosseinMobahi,CeLiu,WilliamT.Freeman
19.APrincipledApproachforCoarse-to-FineMAPInference,ChristopherZach
20.FastApproximateInferenceinHigherOrderMRF-MAPLabelingProblems,ChetanArora,SubhashisBanerjee,PremKalra,S.N.Maheshwari
21.ScanlineSamplerwithoutDetailedBalance:AnEfficientMCMCforMRFOptimization,WonsikKim,KyoungMuLee22.Higher-OrderCliqueReductionWithoutAuxiliaryVariables,HiroshiIshikawa
23.TopicModelingofMultimodalData:AnAutoregressiveApproach,YinZheng,Yu-JinZhang,HugoLarochelle
24.ModelTransport:TowardsScalableTransferLearningonManifolds,OrenFreifeld,SørenHauberg,MichaelJ.Black25.LearningFine-grainedImageSimilaritywithDeep
Ranking,JiangWang,YangSong,ThomasLeung,ChuckRosenberg,JingbinWang,JamesPhilbin,BoChen,YingWu26.AttributedGraphMiningandMatching:AnAttempttoDefineandExtractSoftAttributedPatterns,QuanshiZhang,XuanSong,XiaoweiShao,HuijingZhao,RyosukeShibasaki
27.DeepFisherKernels‒EndtoEndLearningoftheFisherKernelGMMParameters,VladyslavSydorov,MayuSakurada,ChristophH.Lampert
28.TransferJointMatchingforUnsupervisedDomainAdaptation,MingshengLong,JianminWang,GuiguangDing,JiaguangSun,PhilipS.Yu
29.RecognizingRGBImagesbyLearningfromRGB-DData,LinChen,WenLi,DongXu
30.Instance-weightedTransferLearningofActive
AppearanceModels,DanielHaase,ErikRodner,JoachimDenzler

Wednesday,June25(Morning
31.ScalableMultitaskRepresentationLearningforScene
Program
47.GoodVibrations:AModalAnalysisApproachfor
Classification,MaksimLapin,BerntSchiele,MatthiasHein32.LearningtoLearn,fromTransferLearningtoDomainAdaptation:AUnifyingPerspective,NoviPatricia,BarbaraCaputo
33.ConstructingRobustAffinityGraphsforSpectral
Clustering,XiatianZhu,ChenChangeLoy,ShaogangGong34.AFastandRobustAlgorithmtoCountTopologicallyPersistentHolesinNoisyClouds,VitaliyKurlin
35.Co-localizationinReal-WorldImages,KevinTang,ArmandJoulin,Li-JiaLi,LiFei-Fei
36.SpectralClusteringwithJensen-typeKernelsandtheirMulti-pointExtensions,DebarghyaGhoshdastidar,AmbedkarDukkipati,AjayP.Adsul,AparnaS.Vijayan37.FastandRobustArchetypalAnalysisforRepresentationLearning,YuansiChen,JulienMairal,ZaidHarchaoui
38.PhotometricBundleAdjustmentforDenseMulti-View3DModeling,AmaëlDelaunoy,MarcPollefeys
39.ThePhotometryofIntrinsicImages,MarcSerra,OlivierPenacchio,RobertBenavente,MariaVanrell,DimitrisSamaras
40.HighResolution3DShapeTexturefromMultipleVideos,VagiaTsiminaki,Jean-SébastienFranco,EdmondBoyer41.PatchMatchBasedJointViewSelectionandDepthmapEstimation,EnliangZheng,EnriqueDunn,VladimirJojic,Jan-MichaelFrahm
42.LightFieldStereoMatchingUsingBilateralStatisticsofSurfaceCameras,CanChen,HaitingLin,ZhanYu,SingBingKang,JingyiYu
43.RecoveringSurfaceDetailsunderGeneralUnknown
IlluminationUsingShadingandCoarseMulti-viewStereo,DiXu,QiDuan,JianmingZheng,JuyongZhang,JianfeiCai,Tat-JenCham
44.ProbabilisticLabelingCostforHigh-AccuracyMulti-ViewReconstruction,IlyaKostrikov,EstherHorbert,BastianLeibe
45.ComplexNon-RigidMotion3DReconstructionbyUnionofSubspaces,YingyingZhu,DongHuang,FernandoDeLaTorre,SimonLucey
46.AProcrusteanMarkovProcessforNon-RigidStructureRecovery,MinsikLee,Chong-HoChoi,SonghwaiOh
SequentialNon-RigidStructurefromMotion,AntonioAgudo,LourdesAgapito,BegoñaCalvo,JoseM.M.Montiel48.RobustScaleEstimationinReal-TimeMonocularSFMforAutonomousDriving,ShiyuSong,ManmohanChandraker49.OntheQuotientRepresentationfortheEssentialManifold,RobertoTron,KostasDaniilidis
50.EfficientHigh-ResolutionStereoMatchingusingLocalPlaneSweeps,SudiptaN.Sinha,DanielScharstein,RichardSzeliski
51.Cross-ScaleCostAggregationforStereoMatching,KangZhang,YuqiangFang,DongboMin,LifengSun,ShiqiangYang,ShuichengYan,QiTian
52.AsymmetricalGaussMixtureModelsforPointSetsMatching,WenbingTao,KunSun
53.FastandReliableTwo-ViewTranslationEstimation,JohanFredriksson,OlofEnqvist,FredrikKahl
54.GraphCutbasedContinuousStereoMatchingusingLocallySharedLabels,TatsunoriTaniai,YasuyukiMatsushita,TakeshiNaemura
55.LearningtoDetectGroundControlPointsforImprovingtheAccuracyofStereoMatching,AristotleSpyropoulos,NikosKomodakis,PhilipposMordohai
1200–1330Lunch(ExhibitHallC

24


Wednesday,June25(Afternoon
1330-1830PMVideoSpotlights(C213-2151330–1500Special2:PAMI/IJCVSpecialJournal
Session(BattelleGrandSouth
Chairs:RaminZabih(CornellUniv.RonenBasri(WeizmannInst.ofScience
1.2.3.
Program
6.3DPictorialStructuresforMultipleHumanPose
Estimation,VasileiosBelagiannis,SikandarAmin,Mykhaylo
Andriluka,BerntSchiele,NassirNavab,SlobodanIlic
1500–1530Break(BattelleGrandPrefunction1530–1630Oral2E:Face&Gesture
(BattelleGrandSouth
PosterIDsforthissession:O-2E-#where#isthepaper#.Chairs:FernandoDelaTorre(CMURamaChellappa(Univ.ofMaryland
Format(13min.forpresentation+2min.forquestions
1.LearningEuclidean-to-RiemannianMetricforPoint-to-Set
4.5.6.
Format(13min.forpresentation+2min.forquestions
Make3D:Learning3DSceneStructurefromaSingleStillImage,AshutoshSaxena,MinSun,AndrewY.Ng
ProductQuantizationforNearestNeighborSearch,HervéJégou,MatthijsDouze,CordeliaSchmid
ThePASCALVisualObjectClasses(VOCChallenge,MarkEveringham,LucVanGool,ChristopherK.I.Williams,JohnWinn,AndrewZisserman
ConvexandSemi-NonnegativeMatrixFactorizations,ChrisHQDing,TaoLi,MichaelI.Jordan
RobustFaceRecognitionviaSparseRepresentation,JohnWright,AllenY.Yang,ArvindGanesh,S.ShankarSastry,YiMaDeepLearningwithHierarchicalConvolutionalFactor
Analysis,BoChen,GungorPolatkan,GuillermoSapiro,DavidBlei,DavidDunson,LawrenceCarin
1330–1500Oral2D:Attribute-Based
Recognition&HumanPose
Estimation(BattelleGrandNorth
PosterIDsforthissession:O-2D-#where#isthepaper#.Chairs:ChristophLampert(ISTAustriaYiLi(NICTA
1.2.
Classification,ZhiwuHuang,RuipingWang,Shiguang
Shan,XilinChen
2.FaceAlignmentat3000FPSviaRegressingLocalBinaryFeatures,ShaoqingRen,XudongCao,YichenWei,JianSun3.ACompactandDiscriminativeFaceTrackDescriptor,
OmkarM.Parkhi,KarenSimonyan,AndreaVedaldi,AndrewZisserman
4.DeepFace:ClosingtheGaptoHuman-LevelPerformanceinFaceVerification,YanivTaigman,MingYang,Marc'AurelioRanzato,LiorWolf
1530–1630Oral2F:ConvolutionalNeural
Networks(BattelleGrandNorth
PosterIDsforthissession:O-2F-#where#isthepaper#.Chairs:ZhuowenTu(UCSanDiegoFatihPorikli(MERL
Format(13min.forpresentation+2min.forquestions1.FilterForestsforLearningData-DependentConvolutionalKernels,SeanRyanFanello,CemKeskin,PushmeetKohli,ShahramIzadi,JamieShotton,AntonioCriminisi,UgoPattacini,TimPaek
2.LearningandTransferringMid-LevelImage
RepresentationsusingConvolutionalNeuralNetworks,MaximeOquab,LeonBottou,IvanLaptev,JosefSivic
3.Large-scaleVideoClassificationwithConvolutionalNeuralNetworks,AndrejKarpathy,GeorgeToderici,SankethShetty,ThomasLeung,RahulSukthankar,LiFei-Fei
25
3.
4.5.
Format(13min.forpresentation+2min.forquestionsDecorrelatingSemanticVisualAttributesbyResistingtheUrgetoShare,DineshJayaraman,FeiSha,KristenGraumanPANDA:PoseAlignedNetworksforDeepAttributeModeling,NingZhang,ManoharPaluri,Marc'AurelioRanzato,TrevorDarrell,LubomirBourdev
LearningScalableDiscriminativeDictionarywithSampleRelatedness,JiashiFeng,StefanieJegelka,ShuichengYan,TrevorDarrell
DeepPose:HumanPoseEstimationviaDeepNeuralNetworks,AlexanderToshev,ChristianSzegedy
IteratedSecond-OrderLabelSensitivePoolingfor3DHumanPoseEstimation,CatalinIonescu,JoaoCarreira,CristianSminchisescu


Wednesday,June25(Afternoon
4.ConvolutionalNeuralNetworksforNo-ReferenceImage
Program
8.Non-ParametricBayesianConstrainedLocalModels,
QualityAssessment,LeKang,PengYe,YiLi,DavidDoermann
1630–1830Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
1630–1830Demos(C110-115
Real-timeFaceDetectionandRecognitiononGoogleGlass,Shue-Ching,BappadityaMandal,Vijay
Chandrasekhar,ChestonTan,LiyuanLi,JooHweeLim(Inst.forInfocommResearch
RapidandAccurateAvatarCaptureusingaSingleMountedKinect,JongmooChoi,GérardMedioni(USC
PhotoRecall:UsingtheInternettoLabelYourPhotos,NeerajKumar,StevenSeitz(Univ.ofWashington
TheChameleonTrackerin3D,DavidJosephTan,NassirNavab,SlobodanIlic(TechnicalUniv.ofMunich
1630–1830Poster2B:Face&Gesture,
Recognition(GrandBallrooms1-3
PosterIDsforthissession:P-2B-#where#isthepaper#.
1.NonparametricContextModelingofLocalAppearancefor
2.
3.
4.5.
6.7.
Pose-andExpression-RobustFacialLandmark
Localization,BrandonM.Smith,JonathanBrandt,ZheLin,LiZhang
LearningExpressionletsonSpatio-TemporalManifoldforDynamicFacialExpressionRecognition,MengyiLiu,ShiguangShan,RuipingWang,XilinChen
WhoDoILookLike?DeterminingParent-OffspringResemblanceviaGatedAutoencoders,AfshinDehghan,EnriqueG.Ortiz,RubenVillegas,MubarakShah
UnifiedFaceAnalysisbyIterativeMulti-OutputRandomForests,XiaoweiZhao,Tae-KyunKim,WenhanLuo
3
GeometricGenerativeGazeEstimation(GEforRemoteRGB-DCameras,KennethAlbertoFunesMora,Jean-MarcOdobez
AHierarchicalProbabilisticModelforFacialFeatureDetection,YueWu,ZihengWang,QiangJi
RAPS:RobustandEfficientAutomaticConstructionofPerson-SpecificDeformableModels,ChristosSagonas,YannisPanagakis,StefanosZafeiriou,MajaPantic
26
PedroMartins,RuiCaseiro,JorgeBatista
9.FacialExpressionRecognitionviaaBoostedDeepBeliefNetwork,PingLiu,ShizhongHan,ZiboMeng,YanTong10.AutomaticConstructionofDeformableModelsIn-The-Wild,EpameinondasAntonakos,StefanosZafeiriou11.Learning-by-SynthesisforAppearance-based3DGazeEstimation,YusukeSugano,YasuyukiMatsushita,YoichiSato
12.TowardsMulti-viewandPartially-OccludedFaceAlignment,JunliangXing,ZhihengNiu,JunshiHuang,WeimingHu,ShuichengYan
13.HeadPoseEstimationBasedonMultivariateLabelDistribution,XinGeng,YuXia
14.EfficientBoostedExemplar-basedFaceDetection,
HaoxiangLi,ZheLin,JonathanBrandt,XiaohuiShen,GangHua
15.Gauss-NewtonDeformablePartModelsforFaceAlignmentin-the-Wild,GeorgiosTzimiropoulos,MajaPantic
16.IncrementalFaceAlignmentintheWild,AkshayAsthana,StefanosZafeiriou,ShiyangCheng,MajaPantic
17.OneMillisecondFaceAlignmentwithanEnsembleofRegressionTrees,VahidKazemi,JosephineSullivan
18.DiscriminativeDeepMetricLearningforFaceVerificationintheWild,JunlinHu,JiwenLu,Yap-PengTan
19.StackedProgressiveAuto-Encoders(SPAEforFaceRecognitionAcrossPoses,MeinaKan,ShiguangShan,HongChang,XilinChen
20.DeepLearningFaceRepresentationfromPredicting10,000Classes,YiSun,XiaogangWang,XiaoouTang21.OcclusionCoherence:LocalizingOccludedFaceswithaHierarchicalDeformablePartModel,GolnazGhiasi,CharlessC.Fowlkes
22.3D-aidedFaceRecognitionRobusttoExpressionandPoseVariations,BaptisteChu,SamiRomdhani,LimingChen23.LearningNon-LinearReconstructionModelsforImageSetClassification,MunawarHayat,MohammedBennamoun,SenjianAn
24.GestureRecognitionPortfoliosforPersonalization,AngelaYao,LucVanGool,PushmeetKohli


Wednesday,June25(Afternoon
25.SignSpottingusingHierarchicalSequentialPatternswith
Program
40.HierarchicalFeatureHashingforFastDimensionality
TemporalIntervals,Eng-JonOng,OscarKoller,NicolasPugeault,RichardBowden
26.AutomaticFeatureLearningforRobustShadowDetection,SalmanHameedKhan,MohammedBennamoun,FerdousSohel,RobertoTogneri
27.PackingandPadding:CoupledMulti-indexforAccurateImageRetrieval,LiangZheng,ShengjinWang,ZiqiongLiu,QiTian
28.AdaptiveObjectRetrievalwithKernelReconstructiveHashing,HaichuanYang,XiaoBai,JunZhou,PengRen,ZhihongZhang,JianCheng
29.BayesMergingofMultipleVocabulariesforScalableImageRetrieval,LiangZheng,ShengjinWang,WengangZhou,QiTian
30.FastSupervisedHashingwithDecisionTreesforHigh-DimensionalData,GuoshengLin,ChunhuaShen,QinfengShi,AntonvandenHengel,DavidSuter
31.DetectWhatYouCan:DetectingandRepresentingObjectsusingHolisticModelsandBodyParts,XianjieChen,
RoozbehMottaghi,XiaobaiLiu,SanjaFidler,RaquelUrtasun,AlanYuille
32.AssociativeEmbeddingsforLarge-scaleKnowledgeTransferwithSelf-assessment,AlexanderVezhnevets,VittorioFerrari
33.DetectingObjectsusingDeformationDictionaries,BharathHariharan,C.LawrenceZitnick,PiotrDollár
34.Persistence-basedStructuralRecognition,ChunyuanLi,MaksOvsjanikov,FredericChazal
35.InferringUnseenViewsofPeople,Chao-YehChen,KristenGrauman
36.Birdsnap:Large-scaleFine-grainedVisualCategorizationofBirds,ThomasBerg,JiongxinLiu,SeungWooLee,MichelleL.Alexander,DavidW.Jacobs,PeterN.Belhumeur
37.PredictingObjectDynamicsinScenes,DavidF.Fouhey,C.LawrenceZitnick
38.EnrichingVisualKnowledgeBasesviaObjectDiscoveryandSegmentation,XinleiChen,AbhinavShrivastava,AbhinavGupta
39.SeeingtheArrowofTime,LyndseyC.Pickup,ZhengPan,DonglaiWei,YiChangShih,ChangshuiZhang,AndrewZisserman,BernhardSchölkopf,WilliamT.Freeman
27

Reduction,BinZhao,EricP.Xing
41.ModelingImagePatcheswithaGenericDictionaryofMini-Epitomes,GeorgePapandreou,Liang-ChiehChen,AlanL.Yuille
42.Simplex-Based3DSpatio-TemporalFeatureDescriptionforActionRecognition,HaoZhang,WenjunZhou,ChristopherReardon,LynneE.Parker
43.InSearchofInliers:3DCorrespondencebyLocalandGlobalVoting,AndersGlentBuch,YangYang,NorbertKrüger,HenrikGordonPetersen
44.CollectiveMatrixFactorizationHashingforMultimodalData,GuiguangDing,YuchenGuo,JileZhou
45.FindingMatchesinaHaystack:AMax-PoolingStrategyforGraphMatchinginthePresenceofOutliers,MinsuCho,JianSun,OlivierDuchenne,JeanPonce
46.LocalityinGenericInstanceSearchfromOneExample,RanTao,EfstratiosGavves,CeesG.M.Snoek,ArnoldW.M.Smeulders
47.Congruency-BasedReranking,ItaiBen-Shalom,NogaLevy,LiorWolf,NachumDershowitz,AdielBen-Shalom,RoniShweka,YaacovChoueka,TamirHazan,YanivBar48.AsymmetricSparseKernelApproximationsforLarge-scaleVisualSearch,DamekDavis,JonathanBalzer,StefanoSoatto
49.LocallyLinearHashingforExtractingNon-LinearManifolds,GoIrie,ZhenguoLi,Xiao-MingWu,Shih-FuChang
50.ActiveFrame,Location,andDetectorSelectionfor
AutomatedandManualVideoAnnotation,VasiliyKarasev,AvinashRavichandran,StefanoSoatto
51.DistanceEncodedProductQuantization,Jae-PilHeo,ZheLin,Sung-EuiYoon
52.CollaborativeHashing,XianglongLiu,JunfengHe,ChengDeng,BoLang
53.ScalableObjectDetectionusingDeepNeuralNetworks,DumitruErhan,ChristianSzegedy,AlexanderToshev,DragomirAnguelov
1830–2030PAMITCMeeting
(BattelleGrandSouth


Thursday,June26(MorningProgram
0830–1000Oral3B:Video:Events,Activities&
Surveillance(BattelleGrandNorth
PosterIDsforthissession:O-3B-#where#isthepaper#.Chairs:TalHassner(OpenUniv.ofIsraelRahulSukthankar(GoogleResearch
1.
Thursday,June26
0700–1700Registration(ExhibitHallCLobby0730–0830Breakfast(ExhibitHallC
0830-1200AMVideoSpotlights(C213-215
0830–1000Oral3A:Physics-BasedVision&
Shape-from-X(BattelleGrandSouth
PosterIDsforthissession:O-3A-#where#isthepaper#.Chairs:RobertPless(WashingtonUniversityYoichiSato(Univ.ofTokyo
Format(13min.forpresentation+2min.forquestions
1.MultiviewShapeandReflectancefromNatural
2.
3.4.5.
6.
Format(13min.forpresentation+2min.forquestions
Socially-awareLarge-scaleCrowdForecasting,AlexandreAlahi,VigneshRamanathan,LiFei-Fei
L0RegularizedStationaryTimeEstimationforCrowdGroupAnalysis,ShuaiYi,XiaogangWang,CewuLu,JiayaJia
Scene-IndependentGroupProfilinginCrowd,JingShao,ChenChangeLoy,XiaogangWang
TemporalSequenceModelingforVideoEventDetection,YuCheng,QuanfuFan,SharathPankanti,AlokChoudharyRecognitionofComplexEvents:ExploitingTemporalDynamicsbetweenUnderlyingConcepts,SubhabrataBhattacharya,MahdiM.Kalayeh,RahulSukthankar,MubarakShah
VideoEventDetectionbyInferringTemporalInstanceLabels,Kuan-TingLai,FelixX.Yu,Ming-SyanChen,Shih-FuChang
Illumination,GeoffreyOxholm,KoNishino
2.ReflectanceandFluorescentSpectraRecoverybasedon
1000–1030Break(GrandBallroomPrefunction1000–1200Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
3.4.
5.6.
FluorescentChromaticityInvarianceunderVarying
Illumination,YingFu,AntonyLam,YasuyukiKobashi,ImariSato,TakahiroOkabe,YoichiSato
WhatCameraMotionRevealsAboutShapeWithUnknownBRDF,ManmohanChandraker
PhotometricStereousingConstrainedBivariate
RegressionforGeneralIsotropicSurfaces,SatoshiIkehata,KiyoharuAizawa
RobustSeparationofReflectionfromMultipleImages,XiaojieGuo,XiaochunCao,YiMa
Surface-from-Gradients:AnApproachBasedonDiscreteGeometryProcessing,WuyuanXie,YunboZhang,CharlieC.L.Wang,RonaldC.-K.Chung
1000–1200Demos(C110-115
FacialAnalysisforBMIEstimationusingaMobileDevice,YuZhu,LingyunWen,GuodongGuo(WestVirginiaUniv.Real-TimeVideoMagnification,NealWadhwa,MichaelRubinstein,FrédoDurand,WilliamT.Freeman(MITCSAIL,MicrosoftResearch
RealTimeFacialExpressionRecognitiononAndroid,AnkitSharma,OliverNina,LucasPasqualin(Univ.ofCentralFlorida
Ultra-FastAttribute-BasedTransferLearningUsingImagesontheInternet,DaikiKimura,OsamuHasegawa,(TokyoInst.ofTechnology
28



Thursday,June26(Morning
1000–1200Poster3A:Physics-BasedVision,
Recognition,Video:Events,Activities&Surveillance(GrandBallrooms1-3
PosterIDsforthissession:P-3A-#where#isthepaper#.1.BackscatterCompensatedPhotometricStereowith3Sources,ChourmouziosTsiotsios,MariaE.Angelopoulou,Tae-KyunKim,AndrewJ.Davison
2.CalibratingaNon-isotropicNearPointLightSourceusingaPlane,JaesikPark,SudiptaN.Sinha,YasuyukiMatsushita,Yu-WingTai,InSoKweon
3.ANewPerspectiveonMaterialClassificationandInkIdentification,RakeshShiradkar,LiShen,GeorgeLandon,SimHengOng,PingTan
4.HighQualityPhotometricReconstructionusingaDepthCamera,Sk.MohammadulHaque,AvishekChatterjee,VenuMadhavGovindu
5.RobustSurfaceReconstructionviaTripleSparsity,HichamBadri,HusseinYahia,DrissAboutajdine
6.ScatteringParametersandSurfaceNormalsfrom
HomogeneousTranslucentMaterialsusingPhotometricStereo,BoDong,KathleenD.Moore,WeiyiZhang,PieterPeers
7.BetterShadingforBetterShapeRecovery,MoumenT.El-Melegy,AlyS.Abdelrahim,AlyA.Farag
8.StableandInformativeSpectralSignaturesforGraphMatching,NanHu,RaifM.Rustamov,LeonidasGuibas9.DeformableObjectMatchingviaDeformation
Decompositionbased2DLabelMRF,KangweiLiu,JungeZhang,KaiqiHuang,TieniuTan
10.LocallyOptimizedProductQuantizationforApproximateNearestNeighborSearch,YannisKalantidis,YannisAvrithis
11.Multi-sourceDeepLearningforHumanPoseEstimation,WanliOuyang,XiaoChu,XiaogangWang
12.PosebitsforMonocularHumanPoseEstimation,GerardPons-Moll,DavidJ.Fleet,BodoRosenhahn
13.Real-timeSimultaneousPoseandShapeEstimationforArticulatedObjectsUsingaSingleDepthCamera,MaoYe,RuigangYang
Program
14.MixingBody-PartSequencesforHumanPoseEstimation,
AnoopCherian,JulienMairal,KarteekAlahari,CordeliaSchmid
15.RobustEstimationof3DHumanPosesfromaSingle
Image,ChunyuWang,YizhouWang,ZhouchenLin,AlanL.Yuille,WenGao
16.FisherandVLADwithFLAIR,KoenE.A.vandeSande,CeesG.M.Snoek,ArnoldW.M.Smeulders
17.Immediate,ScalableObjectCategoryDetection,YusufAytar,AndrewZisserman
18.WordChannelBasedMultiscalePedestrianDetectionWithoutImageResizingandUsingOnlyOneClassifier,ArthurDanielCostea,SergiuNedevschi
19.ParsingOccludedPeople,GolnazGhiasi,YiYang,DevaRamanan,CharlessC.Fowlkes
20.Multi-foldMILTrainingforWeaklySupervisedObjectLocalization,RamazanGokberkCinbis,JakobVerbeek,CordeliaSchmid
21.GeneratingObjectSegmentationProposalsusingGlobalandLocalSearch,PekkaRantalankila,JuhoKannala,EsaRahtu
22.ANovelChamferTemplateMatchingMethodUsingVariationalMeanField,DucThanhNguyen
23.Confidence-RatedMultipleInstanceBoostingforObjectDetection,KarimAli,KateSaenko
24.COSTA:Co-OccurrenceStatisticsforZero-Shot
Classification,ThomasMensink,EfstratiosGavves,CeesG.M.Snoek
25.AnalysisbySynthesis:3DObjectRecognitionbyObjectReconstruction,MohsenHejrati,DevaRamanan
26.SubmodularObjectRecognition,FanZhu,ZhuolinJiang,LingShao
27.MultimodalLearninginLoosely-organizedWebImages,KunDuan,DavidJ.Crandall,DhruvBatra
28.GeneralizedMaxPooling,NailaMurray,FlorentPerronnin29.DomainAdaptationontheStatisticalManifold,MahsaBaktashmotlagh,MehrtashT.Harandi,BrianC.Lovell,MathieuSalzmann
30.NonparametricPartTransferforFine-grainedRecognition,ChristophGöring,ErikRodner,AlexanderFreytag,JoachimDenzler
29


Thursday,June26(Morning
31.TheFastestDeformablePartModelforObjectDetection,
Program
47.DL-SFA:Deeply-LearnedSlowFeatureAnalysisforAction
JunjieYan,ZhenLei,LongyinWen,StanZ.Li
32.UnsupervisedLearningofDictionariesofHierarchicalCompositionalModels,JifengDai,YiHong,WenzeHu,Song-ChunZhu,YingNianWu
33.QuasiReal-TimeSummarizationforConsumerVideos,BinZhao,EricP.Xing
34.GaitRecognitionunderSpeedTransition,AlMansur,YasushiMakihara,RasyidAqmar,YasushiYagi35.VideoClassificationusingSemanticConceptCo-occurrences,ShayanModiriAssari,AmirRoshanZamir,MubarakShah
36.TemporalSegmentationofEgocentricVideos,YairPoleg,ChetanArora,ShmuelPeleg
37.EfficientActionLocalizationwithApproximately
NormalizedFisherVectors,DanOneata,JakobVerbeek,CordeliaSchmid
38.UnsupervisedTrajectoryModellingusingTemporalInformationviaMinimalPaths,BraisCancela,AlbertoIglesias,MarcosOrtega,ManuelG.Penedo
39.AHierarchicalContextModelforEventRecognitioninSurveillanceVideo,XiaoyangWang,QiangJi40.DISCOVER:DiscoveringImportantSegmentsfor
ClassificationofVideoEventsandRecounting,ChenSun,RamNevatia
41.TowardsGoodPracticesforActionVideoEncoding,JianxinWu,YuZhang,WeiyaoLin
42.ImprovingSemanticConceptDetectionthroughthe
DictionaryofVisually-distinctElements,AfshinDehghan,HaroonIdrees,MubarakShah
43.EfficientFeatureExtraction,EncodingandClassificationforActionRecognition,VadimKantorov,IvanLaptev44.3DPosefromMotionforCross-viewActionRecognitionviaNon-linearCirculantTemporalEncoding,AnkurGupta,JulietaMartinez,JamesJ.Little,RobertJ.Woodham
45.HumanActionRecognitionBasedonContext-DependentGraphKernels,BaoxinWu,ChunfengYuan,WeimingHu46.DepthandSkeletonAssociatedActionRecognition
withoutOnlineAccessibleRGB-DCameras,Yen-YuLin,Ju-HsuanHua,NickC.Tang,Min-HungChen,Hong-YuanMarkLiao
Recognition,LinSun,KuiJia,Tsung-HanChan,YuqiangFang,GangWang,ShuichengYan
48.ACauseandEffectAnalysisofMotionTrajectoriesforModelingActions,SanathNarayan,KalpathiR.Ramakrishnan
49.FromStochasticGrammartoBayesNetwork:ProbabilisticParsingofComplexActivity,NamN.Vo,AaronF.Bobick50.Cross-viewActionModeling,LearningandRecognition,JiangWang,XiaohanNie,YinXia,YingWu,Song-ChunZhu51.VisualSemanticSearch:RetrievingVideosviaComplexTextualQueries,DahuaLin,SanjaFidler,ChenKong,RaquelUrtasun
52.Zero-shotEventDetectionusingMulti-modalFusionofWeaklySupervisedConcepts,ShuangWu,SravanthiBondugula,FlorianLuisier,XiaodanZhuang,PradeepNatarajan
53.DualLinearRegressionBasedClassificationforFaceClusterRecognition,LiangChen
54.BagsofSpacetimeEnergiesforDynamicScene
Recognition,ChristophFeichtenhofer,AxelPinz,RichardP.Wildes
55.Feature-IndependentActionSpottingWithoutHumanLocalization,SegmentationorFrame-wiseTracking,ChuanSun,MarshallTappen,HassanForoosh
1200–1330Lunch(ExhibitHallC
30


Thursday,June26(Afternoon
1330-1830PMVideoSpotlights(C213-2151330–1500Oral3C:Medical&BiologicalImage
Analysis(BattelleGrandSouth
PosterIDsforthissession:O-3C-#where#isthepaper#.Chairs:PetiaRadeva(Univ.ofBarcelonaIoannisKakadiaris(Univ.ofHouston
1.
Program
3.ImageFusionwithLocalSpectralConsistencyand
2.
3.
4.
5.
6.
Format(13min.forpresentation+2min.forquestionsMultiscaleCenterlineDetectionbyLearningaScale-SpaceDistanceTransform,AmosSironi,VincentLepetit,PascalFua
MultivariateGeneralLinearModels(MGLMon
RiemannianManifoldswithApplicationstoStatisticalAnalysisofDiffusionWeightedImages,HyunwooJ.Kim,NageshAdluru,MaxwellD.Collins,MooK.Chung,BarbaraB.Bendlin,SterlingC.Johnson,RichardJ.Davidson,VikasSingh
PreconditioningforAcceleratedIterativelyReweightedLeastSquaresinStructuredSparsityReconstruction,ChenChen,JunzhouHuang,LeiHe,HongshengLi
JointCoupled-FeatureRepresentationandCoupledBoostingforADDiagnosis,YinghuanShi,Heung-IlSuk,YangGao,DinggangShen
DeformableRegistrationofFeature-EndowedPointSetsBasedonTensorFields,DemianWassermann,JamesRoss,GeorgeWashko,WilliamM.WellsIII,RaulSanJose-EsteparTrackingIndistinguishableTranslucentObjectsoverTimeusingWeaklySupervisedStructuredLearning,LucaFiaschi,FerranDiego,KonstantinGregor,MartinSchiegg,UllrichKoethe,MartaZlatic,FredA.Hamprecht
DynamicGradientSparsity,ChenChen,YeqingLi,WeiLiu,JunzhouHuang
4.Segmentation-FreeDynamicSceneDeblurring,TaeHyunKim,KyoungMuLee
5.ShrinkageFieldsforEffectiveImageRestoration,UweSchmidt,StefanRoth
6.CamouflaginganObjectfromManyViewpoints,AndrewOwens,ConnellyBarnes,AlexFlint,HanumantSingh,WilliamFreeman
1500–1530Break(BattelleGrandPrefunction1530–1630Special3:PlenarySession
(BattelleGrand
Chairs:ReneVidal(JohnsHopkinsUniv.RonenBasri(WeizmannInst.ofSciencePlenaryTalk:AreDeepNetworksaSolutiontotheCurseofDimensionality?,StéphaneMallat(ÉcoleNormaleSupérieure
Abstract:Learninggaveaconsiderableandsurprisingboosttocomputervision,anddeepneuralnetworksappeartobethenewwinnersofthefierceraceon
classificationerrors.Algorithmrefinementsarenowgoingwellbeyondourunderstandingoftheproblem,andseemtomakeirrelevantanystudyofcomputervisionmodels.Yet,learningfromhigh-dimensionaldatasuchasimages,suffersfromacurseofdimensionalitywhichpredictsacombinatorialexplosion.Whyaretheseneuralarchitecturesavoidingthiscurse?Isthisrootedinpropertiesofimagesandvisualtasks?Canthese
propertiesberelatedtohigh-dimensionalproblemsinotherfields?Weshallexplorethemathematicalrootsofthesequestions,andtellastorywhereinvariants,
contractions,sparsity,dimensionreductionandmultiscaleanalysisplayimportantroles.Imagesandexampleswillgiveacolorfulbackgroundtothetalk.
1330–1500Oral3D:Low-LevelVision&Image
Processing(BattelleGrandNorth
PosterIDsforthissession:O-3D-#where#isthepaper#.Chairs:MichaelBrown(NationalUniv.ofSingaporeStefanRoth(TUDarmstadtFormat(13min.forpresentation+2min.forquestions1.Scale-spaceProcessingUsingPolynomial
Representations,GouKoutaki,KeiichiUchimura
2.SingleImageLayerSeparationusingRelativeSmoothness,YuLi,MichaelS.Brown
31

1630–1830Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
1630–1830Demos(C110-115
PanOptus:AutomaticVideoEditingforiPhone,GoogleGlassandSurveillanceCamera,BinZhao,BinShu,EricXing(CMU

Thursday,June26(Afternoon
Surface-from-Gradient(SfGbyDiscreteGeometryProcessing(DGP,WuyuanXie,YunboZhang,HarlieC.L.Wang,RonaldC.-K.Chung(ChineseUniv.ofHongKongComplexActivityDetectionandFunctionalSceneUnderstandinginVideo,EranSwears,AnthonyHoogs,SangminOh,MattLeotta(KitwareInc.
CloudCV:LargeScaleDistributedComputerVisionasaCloudService,HarshAgrawal,NeelimaChavali,ClintSolomonMathialagan,AbdullahAlfadda,PrakritiBanik,DhruvBatra(VirginiaTech
Program
12.BlindImageQualityAssessmentusingSemi-supervised
1630–1830Poster3B:BiologicallyInspired
Vision,Low-LevelVision,Medical&BiologicalImageAnalysis,
Segmentation(GrandBallrooms1-3
PosterIDsforthissession:P-3B-#where#isthepaper#.
1.LearningOptimalSeedsforDiffusion-basedSalientObject
Detection,SongLu,VijayMahadevan,NunoVasconcelos2.Large-ScaleOptimizationofHierarchicalFeaturesforSaliencyPredictioninNaturalImages,EleonoraVig,MichaelDorr,DavidCox
3.SaliencyDetectiononLightField,NianyiLi,JinweiYe,YuJi,HaibinLing,JingyiYu
4.SaliencyOptimizationfromRobustBackgroundDetection,WangjiangZhu,ShuangLiang,YichenWei,JianSun
5.AReverseHierarchyModelforPredictingEyeFixations,TianlinShi,MingLiang,XiaolinHu
6.100+TimesFasterWeightedMedianFilter(WMF,QiZhang,LiXu,JiayaJia
7.Edge-awareGradientDomainOptimizationFrameworkforImageFilteringbyLocalPropagation,MiaoHua,XiaohuiBie,MinyingZhang,WenchengWang
8.Super-ResolvingNoisyImages,AbhishekSingh,FatihPorikli,NarendraAhuja
9.SparseDictionaryLearningforEditPropagationofHigh-ResolutionImages,XiaowuChen,DongqingZou,JianweiLi,XiaochunCao,QinpingZhao,HaoZhang
10.WeightedNuclearNormMinimizationwithApplicationtoImageDenoising,ShuhangGu,LeiZhang,WangmengZuo,XiangchuFeng
11.UsingProjectionKurtosisConcentrationOfNatural
ImagesForBlindNoiseCovarianceMatrixEstimation,XingZhang,SiweiLyu
32

RectifierNetworks,HuixuanTang,NeelJoshi,AshishKapoor
13.SeparableKernelforImageDeblurring,LuFang,HaifengLiu,FengWu,XiaoyanSun,HouqiangLi
14.JointDepthEstimationandCameraShakeRemovalfromSingleBlurryImage,ZheHu,LiXu,Ming-HsuanYang15.DeblurringTextImagesviaL0-RegularizedIntensityandGradientPrior,JinshanPan,ZheHu,ZhixunSu,Ming-HsuanYang
16.TotalVariationBlindDeconvolution:TheDevilisintheDetails,DanielePerrone,PaoloFavaro
17.SingleImageSuper-resolutionusingDeformablePatches,YuZhu,YanningZhang,AlanL.Yuille
18.Multi-ShotImaging:JointAlignment,Deblurringand
Resolution-Enhancement,HaichaoZhang,LawrenceCarin19.CID:CombinedImageDenoisinginSpatialandFrequencyDomainsUsingWebImages,HuanjingYue,XiaoyanSun,JingyuYang,FengWu
20.MultipointFilteringwithLocalPolynomialApproximationandRangeGuidance,XiaoTan,ChangmingSun,TuanD.Pham
21.DecomposableNonlocalTensorDictionaryLearningforMultispectralImageDenoising,YiPeng,DeyuMeng,ZongbenXu,ChenqiangGao,YiYang,BiaoZhang22.Robust3DFeaturesforMatchingbetweenDistorted
RangeScansCapturedbyMovingSystems,XiangqiHuang,BoZheng,TakeshiMasuda,KatsushiIkeuchi
23.DiscriminativeBlurDetectionFeatures,JianpingShi,LiXu,JiayaJia
24.Detection,RectificationandSegmentationofCoplanarRepeatedPatterns,JamesPritts,OndřejChum,JiříMatas25.MirrorSymmetryHistogramsforCapturingGeometricPropertiesinImages,MarceloCicconet,DaviGeiger,KristinCGunsalus,MichaelWerman
26.ALearning-to-RankApproachforImageColor
Enhancement,JianzhouYan,StephenLin,SingBingKang,XiaoouTang
27.InvestigatingHaze-relevantFeaturesinALearningFrameworkforImageDehazing,KetanTang,JianchaoYang,JueWang

Thursday,June26(Afternoon
28.QualityAssessmentforComparingImageEnhancement
Program
42.Learning-BasedAtlasSelectionforMultiple-Atlas
Algorithms,ZhengyingChen,TingtingJiang,YonghongTian29.ShadowRemovalfromSingleRGB-DImages,YaoXiao,EfstratiosTsougenis,Chi-KeungTang
30.ManifoldBasedDynamicTextureSynthesisfrom
ExtremelyFewSamples,HongtengXu,HongyuanZha,MarkA.Davenport
31.TheSynthesizabilityofTextureExamples,DengxinDai,HaykoRiemenschneider,LucVanGool
32.ReconstructingEvolvingTreeStructuresinTimeLapseSequences,PrzemysławGłowacki,MiguelAmávelPinheiro,EnginTüretken,RaphaelSznitman,DanielLebrecht,JanKybic,AnthonyHoltmaat,PascalFua
33.Total-VariationMinimizationonUnstructuredVolumetricMesh:BiophysicalApplicationsonReconstructionof3DIschemicMyocardium,JingjiaXu,AzarRahimiDehaghani,FeiGao,LinweiWang
34.TrackingontheProductManifoldofShapeandOrientationforTractographyfromDiffusionMRI,YuanxiangWang,HesamoddinSalehian,GuangCheng,BabaC.Vemuri
35.CurvilinearStructureTrackingbyLowRankTensor
ApproximationwithModelPropagation,ErkangCheng,YuPang,YingZhu,JingyiYu,HaibinLing
36.Patch-basedEvaluationofImageSegmentation,ChristianLedig,WenzheShi,WenjiaBai,DanielRueckert
37.EvaluationofScan-LineOptimizationfor3DMedicalImageRegistration,SimonHermann
38.ClassificationofHistologySectionsviaMultispectralConvolutionalSparseCoding,YinZhou,HangChang,KennethBarner,PaulSpellman,BahramParvin39.Matrix-SimilarityBasedLossFunctionandFeature
SelectionforAlzheimer'sDiseaseDiagnosis,XiaofengZhu,Heung-IlSuk,DinggangShen
40.DiscriminativeSparseInverseCovarianceMatrix:
ApplicationinBrainFunctionalNetworkClassification,LupingZhou,LeiWang,PhilipOgunbona
41.ABayesianFrameworkFortheLocalConfigurationofRetinalJunctions,TouseefAhmadQureshi,AndrewHunter,BashirAl-Diri
Segmentation,GerardSanroma,GuorongWu,YaozongGao,DinggangShen
43.FullyAutomatedNon-rigidSegmentationwithDistanceRegularizedLevelSetEvolutionInitializedand
ConstrainedbyDeep-structuredInference,TuanAnhNgo,GustavoCarneiro
44.FASTLABEL:EasyandEfficientSolutionofJointMulti-LabelandEstimationProblems,GaneshSundaramoorthi,Byung-WooHong
45.LearningtoGroupObjects,VictoriaYanulevskaya,JasperUijlings,NicuSebe
46.UnsupervisedMulti-ClassJointImageSegmentation,FanWang,QixingHuang,MaksOvsjanikov,LeonidasJ.Guibas47.SemanticObjectSelection,EjazAhmed,ScottCohen,BrianPrice
48.Discrete-ContinuousGradientOrientationEstimationforFasterImageSegmentation,MichaelDonoser,DieterSchmalstieg
49.Object-basedMultipleForegroundVideoCo-segmentation,HuazhuFu,DongXu,BaoZhang,StephenLin
50.ParsingWorld'sSkylinesusingShape-ConstrainedMRFs,RashmiTonge,SubhransuMaji,C.V.Jawahar
51.ClothingCo-ParsingbyJointImageSegmentationandLabeling,WeiYang,PingLuo,LiangLin
52.TellMeWhatYouSeeandIwillShowYouWhereItIs,JiaXu,AlexanderG.Schwing,RaquelUrtasun
53.BeattheMTurkers:AutomaticImageLabelingfromWeak3DSupervision,Liang-ChiehChen,SanjaFidler,AlanL.Yuille,RaquelUrtasun
54.EfficientStructuredParsingofFaçadesUsingDynamicProgramming,AndreaCohen,AlexanderG.Schwing,MarcPollefeys
55.DenseSemanticImageSegmentationwithObjectsandAttributes,ShuaiZheng,Ming-MingCheng,Jonathan
Warrell,PaulSturgess,VibhavVineet,CarstenRother,PhilipH.S.Torr
1830–2030Reception(BattelleGrand
33


Friday,June27(MorningProgram
0830–1000Oral4B:Recognition:Detection,
Categorization,Classification(BattelleGrandNorth
PosterIDsforthissession:O-4B-#where#isthepaper#.Chairs:HaibinLing(TempleUniv.JiriMatas(CzechTechnicalUniv.
1.
Friday,June27
0700–1700Registration(ExhibitHallCLobby0730–0830Breakfast(ExhibitHallC
0830-1200AMVideoSpotlights(C213-215
0830–1000Oral4A:Computational
Photography:SensingandDisplay(BattelleGrandSouth
PosterIDsforthissession:O-4A-#where#isthepaper#.Chairs:HuiJi(NationalUniv.ofSingaporeKyrosKutulakos(Univ.ofToronto
1.
2.3.
4.
5.6.
2.
3.
4.
5.6.
Format(13min.forpresentation+2min.forquestionsDiffuseMirrors:3DReconstructionfromDiffuseIndirectIlluminationUsingInexpensiveTime-of-FlightSensors,FelixHeide,LeiXiao,WolfgangHeidrich,MatthiasB.HullinFourierAnalysisonTransientImagingwitha
MultifrequencyTime-of-FlightCamera,JingyuLin,YebinLiu,MatthiasB.Hullin,QionghaiDai
TransparentObjectReconstructionviaCodedTransportofIntensity,ChenguangMa,XingLin,JinliSuo,QionghaiDai,GordonWetzstein
3DShapeandIndirectAppearancebyStructuredLightTransport,MatthewO'Toole,JohnMather,KiriakosN.Kutulakos
Shape-PreservingHalf-ProjectiveWarpsforImageStitching,Che-HanChang,YoichiSato,Yung-YuChuangParallax-tolerantImageStitching,FanZhang,FengLiu
Format(13min.forpresentation+2min.forquestionsLearningEverythingaboutAnything:Webly-SupervisedVisualConceptLearning,SantoshK.Divvala,AliFarhadi,CarlosGuestrin
Dirichlet-basedHistogramFeatureTransformforImageClassification,TakumiKobayashi
BING:BinarizedNormedGradientsforObjectnessEstimationat300fps,Ming-MingCheng,ZimingZhang,Wen-YanLin,PhilipTorr
ContextDrivenSceneParsingwithAttentiontoRareClasses,JimeiYang,BrianPrice,ScottCohen,Ming-HsuanYang
PatchtotheFuture:UnsupervisedVisualPrediction,JacobWalker,AbhinavGupta,MartialHebert
TriangulationEmbeddingandDemocraticAggregationforImageSearch,HervéJégou,AndrewZisserman
1000–1030Break(GrandBallroomPrefunction1000–1200Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
1000–1200Demos(C110-115
MobileVisionforCommunityCare:AFrameworkfor
PeoplewithDementia,MoiHoonYap,Choon-ChingNg,BillCassidy,GemmaStringer(ManchesterMetropolitanUniv.Large-Scale,RealTimeVisual-InertialNavigationandMapping,KonstantineTsotsos,StephenPhillips,StefanoSoatto(UCLA


34


Friday,June27(MorningProgram
17.HumanShapeandPoseTrackingUsingKeyframes,Chun-
1000–1200Poster4A:Computational
Photography,Motion&Tracking,Recognition(GrandBallrooms1-3
PosterIDsforthissession:P-4A-#where#isthepaper#.
1.Low-CostCompressiveSensingforColorVideoandDepth,
XinYuan,PatrickLlull,XuejunLiao,JianboYang,DavidJ.Brady,GuillermoSapiro,LawrenceCarin
2.AliasingDetectionandReductioninPlenopticImaging,ZhaolinXiao,QingWang,GuoqingZhou,JingyiYu
3.Illumination-AwareAgeProgression,IraKemelmacher-Shlizerman,SupasornSuwajanakorn,StevenM.Seitz4.ColorTransferUsingProbabilisticMovingLeastSquares,YoungbaeHwang,Joon-YoungLee,InSoKweon,SeonJooKim
5.ImagePre-compensation:BalancingContrastandRinging,YuJi,JinweiYe,SingBingKang,JingyiYu
6.Time-MappingUsingSpace-TimeSaliency,FengZhou,SingBingKang,MichaelF.Cohen
7.Gyro-BasedMulti-ImageDeconvolutionforRemovingHandshakeBlur,SungHeePark,MarcLevoy
8.Similarity-AwarePatchworkAssemblyforDepthImageSuper-Resolution,JingLi,ZhichaoLu,GangZeng,RuiGan,HongbinZha
9.DeblurringLow-lightImageswithLightStreaks,ZheHu,SunghyunCho,JueWang,Ming-HsuanYang
10.DepthEnhancementviaLow-rankMatrixCompletion,SiLu,XiaofengRen,FengLiu
11.Raw-to-Raw:MappingbetweenImageSensorColor
Responses,RangNguyen,DilipK.Prasad,MichaelS.Brown12.DAISYFilterFlow:AGeneralizedDiscreteApproachtoDenseCorrespondences,HongshengYang,Wen-YanLin,JiangboLu
13.Robust3DTrackingwithDescriptorFields,AlbertoCrivellaro,VincentLepetit
14.EvolutionaryQuasi-randomSearchforHandArticulationsTracking,IasonOikonomidis,ManolisI.A.Lourakis,AntonisA.Argyros
15.Scalable3DTrackingofMultipleInteractingObjects,NikolaosKyriazis,AntonisArgyros
16.BayesianActiveAppearanceModels,JoanAlabort-i-Medina,StefanosZafeiriou
35

HaoHuang,EdmondBoyer,NassirNavab,SlobodanIlic18.BetterFeatureTrackingThroughSubspaceConstraints,BryanPoling,GiladLerman,ArthurSzlam
19.GroundPlaneEstimationusingaHiddenMarkovModel,RalfDragon,LucVanGool
20.OnlineObjectTracking,LearningandParsingwithAnd-OrGraphs,YangLu,TianfuWu,SongChunZhu21.Region-basedParticleFilterforVideoObjectSegmentation,DavidVaras,FerranMarques
22.VisualTrackingviaProbabilityContinuousOutlierModel,DongWang,HuchuanLu
23.VisualTrackingUsingPertinentPatchSelectionandMasking,Dae-YounLee,Jae-YoungSim,Chang-SuKim24.IntervalTracker:TrackingbyIntervalAnalysis,JunseokKwon,KyoungMuLee
25.UnifyingSpatialandAttributeSelectionforDistracter-ResilientTracking,NanJiang,YingWu
26.PedestrianDetectioninLow-resolutionImageryby
LearningMulti-scaleIntrinsicMotionStructures(MIMS,JiejieZhu,OmarJaved,JingenLiu,QianYu,HuiCheng,HarpreetSawhney
27.Multi-targetTrackingwithMotionContextinTensorPowerIteration,XinchuShi,HaibinLing,WeimingHu,ChunfengYuan,JunliangXing
28.SphereFlow:6DoFSceneFlowfromRGB-DPairs,MichaelHornáček,AndrewFitzgibbon,CarstenRother
29.FastEdge-PreservingPatchMatchforLargeDisplacementOpticalFlow,LinchaoBao,QingxiongYang,HailinJin30.LearninganImage-basedMotionContextforMultiplePeopleTracking,LauraLeal-Taixé,MicheleFenzi,AlinaKuznetsova,BodoRosenhahn,SilvioSavarese
31.Semi-SupervisedCoupledDictionaryLearningforPersonRe-identification,XiaoLiu,MingliSong,DachengTao,XingchenZhou,ChunChen,JiajunBu
32.WhatareYouTalkingAbout?Text-to-ImageCoreference,ChenKong,DahuaLin,MohitBansal,RaquelUrtasun,SanjaFidler
33.PredictingFailuresofVisionSystems,PengZhang,JiulingWang,AliFarhadi,MartialHebert,DeviParikh

Friday,June27(MorningProgram
48.2DHumanPoseEstimation:NewBenchmarkandStateof
34.ThreeGuidelinesofOnlineLearningforLarge-ScaleVisual
Recognition,YoshitakaUshiku,MasatoshiHidaka,TatsuyaHarada
35.Usingk-PoseletsforDetectingPeopleandLocalizingTheirKeypoints,GeorgiaGkioxari,BharathHariharan,RossGirshick,JitendraMalik
36.RandomizedMax-MarginCompositionsforVisual
Recognition,AngelaEigenstetter,MasatoTakami,BjörnOmmer
37.Large-ScaleVisualFontRecognition,GuangChen,
JianchaoYang,HailinJin,JonathanBrandt,EliShechtman,AseemAgarwala,TonyX.Han
38.DescribingTexturesintheWild,MirceaCimpoi,SubhransuMaji,IasonasKokkinos,SammyMohamed,AndreaVedaldi39.RelativeParts:DistinctivePartsforLearningRelativeAttributes,RamachandruniN.Sandeep,YashaswiVerma,C.V.Jawahar
40.UnderstandingObjectsinDetailwithFine-Grained
Attributes,AndreaVedaldi,SiddharthMahendran,StavrosTsogkas,SubhransuMaji,RossGirshick,JuhoKannala,EsaRahtu,IasonasKokkinos,MatthewB.Blaschko,David
Weiss,BenTaskar,KarenSimonyan,NaomiSaphra,SammyMohamed
41.PredictingUserAnnoyanceUsingVisualAttributes,GordonChristie,AmarParkash,UjwalKrothapalli,DeviParikh
42.LinearRankingAnalysis,WeihongDeng,JianiHu,JunGuo43.TransformationPursuitforImageClassification,MattisPaulin,JérômeRevaud,ZaidHarchaoui,FlorentPerronnin,CordeliaSchmid
44.IncrementalLearningofNCMForestsforLarge-ScaleImageClassification,MarkoRistin,MatthieuGuillaumin,JuergenGall,LucVanGool
45.ObjectClassificationwithAdaptableRegions,HakanBilen,MarcoPedersoli,VinayP.Namboodiri,TinneTuytelaars,LucVanGool
46.DiscriminativeFernsEnsembleforHandPoseRecognition,EyalKrupka,AlonVinnikov,BenKlein,AharonBarHillel,DanielFreedman,SimonStachniak
47.AreCarsJust3DBoxes?‒JointlyEstimatingthe3DShapeofMultipleObjects,MuhammadZeeshanZia,MichaelStark,KonradSchindler
36

theArtAnalysis,MykhayloAndriluka,LeonidPishchulin,PeterGehler,BerntSchiele
49.UsingaDeformationFieldModelforLocalizingFacesandFacialPointsunderWeakSupervision,MarcoPedersoli,TinneTuytelaars,LucVanGool
50.ActiveAnnotationTranslation,SteveBranson,KristjánEldjárnHjörleifsson,PietroPerona
51.LookingBeyondtheVisibleScene,AdityaKhosla,ByoungkwonAnAn,JosephJ.Lim,AntonioTorralba
52.Two-ClassWeatherClassification,CewuLu,DiLin,JiayaJia,Chi-KeungTang
53.LearningImportantSpatialPoolingRegionsforSceneClassification,DiLin,CewuLu,RenjieLiao,JiayaJia
54.OrientationalPyramidMatchingforRecognizingIndoorScenes,LingxiXie,JingdongWang,BainingGuo,BoZhang,QiTian
55.MultilabelRankingwithInconsistentRankers,XinGeng,LongrunLuo
56.SceneParsingwithObjectInstancesandOcclusionOrdering,JosephTighe,MarcNiethammer,SvetlanaLazebnik
1200–1330Lunch(ExhibitHallC

Friday,June27(Afternoon
1330-1830PMVideoSpotlights(C213-215
1330–1500Oral4C:3DGeometry&Shape
(BattelleGrandSouth
Program
1330–1500Oral4D:StatisticalMethodsand
LearningII(BattelleGrandNorth
PosterIDsforthissession:O-4D-#where#isthepaper#.Chairs:JasonCorso(SUNYBuffaloZhouchenLin(PekingUniv.
1.
PosterIDsforthissession:O-4C-#where#isthepaper#.Chairs:LourdesAgapito(Univ.CollegeLondon
PaoloFavaro(Univ.ofBern
1.
2.
3.
4.5.
6.
Format(13min.forpresentation+2min.forquestionsARiemannianFrameworkforMatchingPointClouds
RepresentedbytheSchrödingerDistanceTransform,YanDeng,AnandRangarajan,StephanEisenschenk,BabaC.Vemuri
Seeing3DChairs:ExemplarPart-based2D-3DAlignmentusingaLargeDatasetofCADModels,MathieuAubry,DanielMaturana,AlexeiA.Efros,BryanC.Russell,JosefSivic
AMixtureofManhattanFrames:BeyondtheManhattanWorld,JulianStraub,GuyRosman,OrenFreifeld,JohnJ.Leonard,JohnW.FisherIII
LocalRegularity-drivenCity-scaleFacadeDetectionfromAerialImages,JingchenLiu,YanxiLiu
LatentRegressionForest:StructuredEstimationof3DArticulatedHandPosture,DanhangTang,HyungJinChang,AlykhanTejani,Tae-KyunKim
FAUST:DatasetandEvaluationfor3DMeshRegistration,FedericaBogo,JavierRomero,MatthewLoper,MichaelJ.Black
2.3.4.5.6.
Format(13min.forpresentation+2min.forquestionsOptimizingOverRadialKernelsonCompactManifolds,SadeepJayasumana,RichardHartley,MathieuSalzmann,HongdongLi,MehrtashHarandi
GrassmannAveragesforScalableRobustPCA,SørenHauberg,AasaFeragen,MichaelJ.Black
RobustSubspaceSegmentationwithBlock-diagonalPrior,JiashiFeng,ZhouchenLin,HuanXu,ShuichengYan
UnsupervisedOne-ClassLearningforAutomaticOutlierRemoval,WeiLiu,GangHua,JohnR.Smith
SmoothRepresentationClustering,HanHu,ZhouchenLin,JianjiangFeng,JieZhou
NovelMethodsforMultilinearDataCompletionandDe-noisingBasedonTensor-SVD,ZeminZhang,GregoryEly,ShuchinAeron,NingHao,MishaKilmer
1500–1530Break(BattelleGrandPrefunction1530–1630Oral4E:OptimizationMethods
(BattelleGrandSouth
PosterIDsforthissession:O-4E-#where#isthepaper#.Chairs:AllenYang(UCBerkeleyYiMa(ShanghaiTechUniv.&UIUC
1.2.

3.
4.
Format(13min.forpresentation+2min.forquestionsSecond-OrderShapeOptimizationforGeometricInverseProblemsinVision,JonathanBalzer,StefanoSoatto
0NormBasedDictionaryLearningbyProximalMethodswithGlobalConvergence,ChenglongBao,HuiJi,YuhuiQuan,ZuoweiShen
AdaptivePartialDifferentialEquationLearningforVisualSaliencyDetection,RishengLiu,JunjieCao,ZhouchenLin,ShiguangShan
RobustOrthonormalSubspaceLearning:Efficient
RecoveryofCorruptedLow-rankMatrices,XianbiaoShu,FatihPorikli,NarendraAhuja
37


Friday,June27(AfternoonProgram
5.TurningMobilePhonesinto3DScanners,KalinKolev,Petri
1530–1630Oral4F:ViewSynthesis&Other
Applications(BattelleGrandNorth
PosterIDsforthissession:O-4F-#where#isthepaper#.Chairs:KostasDaniilidis(Univ.ofPennsylvaniaGérardMedioni(USC
1.
2.
3.
4.
Format(13min.forpresentation+2min.forquestionsReconstructingStorylineGraphsforImage
RecommendationfromWebCommunityPhotos,GunheeKim,EricP.Xing
ActiveFlatteningofCurvedDocumentImagesviaTwoStructuredBeams,GaofengMeng,YingWang,ShenquanQu,ShimingXiang,ChunhongPan
Image-basedSynthesisandRe-SynthesisofViewpointsGuidedby3DModels,KonstantinosRematas,TobiasRitschel,MarioFritz,TinneTuytelaars
BayesianViewSynthesisandImage-BasedRenderingPrinciples,SergiPujades,FrédéricDevernay,BastianGoldluecke
1630–1830Exhibits(GrandBallrooms1-3
SameasTuesdaymorningExhibits(seepg.15
1630–1830Poster4B:3DVision,Document
Analysis,OptimizationMethods,Shape,VisionforGraphics,Web&VisionSystems(GrandBallrooms1-3
PosterIDsforthissession:P-4B-#where#isthepaper#.1.FastMRFOptimizationwithApplicationtoDepthReconstruction,QifengChen,VladlenKoltun
2.ExploitingShadingCuesinKinectIRImagesforGeometryRefinement,GyeongminChoe,JaesikPark,Yu-WingTai,InSoKweon
3.FastRotationSearchwithStereographicProjectionsfor3DRegistration,ÁlvaroParraBustos,Tat-JunChin,DavidSuter
4.LocalReadjustmentforHigh-Resolution3D
Reconstruction,SiyuZhu,TianFang,JianxiongXiao,LongQuan
Tanskanen,PabloSpeciale,MarcPollefeys
6.T-Linkage:AContinuousRelaxationofJ-LinkageforMulti-ModelFitting,LucaMagri,AndreaFusiello
7.Motion-Depth:RGB-DDepthMapEnhancementwithMotionandDepthinComplement,Tak-WaiHui,KingNgiNgan
8.GeneralizedPupil-CentricImagingandAnalyticalCalibrationforaNon-frontalCamera,AvinashKumar,NarendraAhuja
9.GeometricUrbanGeo-Localization,MayankBansal,KostasDaniilidis
10.3DReconstructionfromAccidentalMotion,FisherYu,DavidGallup
11.Real-timeModel-basedArticulatedObjectPoseDetectionandTrackingwithVariableRigidityConstraints,KarlPauwels,LeonardoRubio,EduardoRos
12.OccludingContoursforMulti-ViewStereo,QiShan,BrianCurless,YasutakaFurukawa,CarlosHernandez,StevenM.Seitz
13.AerialReconstructionsviaProbabilisticDataFusion,RandiCabezas,OrenFreifeld,GuyRosman,JohnW.FisherIII14.3DModelingfromWideBaselineRangeScansusingContourCoherence,RuizheWang,JongmooChoi,GérardMedioni
15.OrientationRobustTextLineDetectioninNaturalImages,LeKang,YiLi,David,Doermann
16.Strokelets:ALearnedMulti-ScaleRepresentationforSceneTextRecognition,CongYao,XiangBai,BaoguangShi,WenyuLiu
17.Region-basedDiscriminativeFeaturePoolingforSceneTextRecognition,Chen-YuLee,AnuragBhardwaj,WeiDi,VigneshJagadeesh,RobinsonPiramuthu
18.FastandExact:ADMM-BasedDiscriminativeShape
SegmentationwithLoopyPartModels,HaithemBoussaid,IasonasKokkinos
19.PseudoconvexProximalSplittingforLProblemsinMultiviewGeometry,AndersEriksson,MatsIsaksson20.AConvexRelaxationoftheAmbrosio—TortorelliEllipticFunctionalsfortheMumford-ShahFunctional,YoungwookKee,JunmoKim
38


Friday,June27(AfternoonProgram
36.Symmetry-AwareNonrigidMatchingofIncomplete3D
21.MultiLabelGenericCuts:OptimalInferenceinMultiLabel
MultiCliqueMRF-MAPProblems,ChetanArora,S.N.Maheshwari
22.SequentialConvexRelaxationforMutualInformation-BasedUnsupervisedFigure-GroundSegmentation,
YoungwookKee,MohamedSouiai,DanielCremers,JunmoKim
23.DecorrelatedVectorialTotalVariation,ShunsukeOno,IsaoYamada
24.EfficientSquaredCurvature,ClaudiaNieuwenhuis,EnoToeppe,LenaGorelick,OlgaVeksler,YuriBoykov25.Multi-featureSpectralClusteringwithMinimax
Optimization,HongxingWang,ChaoqunWeng,JunsongYuan
26.Quality-basedMultimodalClassificationusingTree-StructuredSparsity,SoheilBahrampour,AsokRay,NasserM.Nasrabadi,KennethW.Jenkins
27.NewtonGreedyPursuit:AQuadraticApproximation
MethodforSparsity-ConstrainedOptimization,Xiao-TongYuan,QingshanLiu
28.GeneralizedNonconvexNonsmoothLow-RankMinimization,CanyiLu,JinhuiTang,ShuichengYan,ZhouchenLin
29.LatentDictionaryLearningforSparseRepresentationbasedClassification,MengYang,DengxinDai,LilinShen,LucVanGool
30.IsRotationaNuisanceinShapeRecognition?,QiuhongKe,YiLi
31.Dual-SpaceDecompositionof2DComplexShapes,GuilinLiu,ZhonghuaXi,Jyh-MingLien
32.NoisingversusSmoothingforVertexIdentificationinUnknownShapes,KonstantinosA.Raftopoulos,MarinFerecatu
33.SurfaceRegistrationbyOptimizationinConstrainedDiffeomorphismSpace,WeiZeng,LokMingLui,XianfengGu
34.DenseNon-RigidShapeCorrespondenceusingRandomForests,EmanueleRodolà,SamuelRotaBulò,ThomasWindheuser,MatthiasVestner,DanielCremers
35.CovarianceDescriptorsfor3DShapeMatchingand
Retrieval,HediTabia,HamidLaga,DavidPicard,Philippe-HenriGosselin
39

Surfaces,YusukeYoshiyasu,EiichiYoshida,KazuhitoYokoi,RyusukeSagawa
37.AnAutomatedEstimatorofImageVisualRealismBasedonHumanCognition,ShaojingFan,Tian-TsongNg,
JonathanS.Herberg,BryanL.Koenig,ChestonY.-C.Tan,RangdingWang
38.SteadyFlow:SpatiallySmoothOpticalFlowforVideoStabilization,ShuaichengLiu,LuYuan,PingTan,JianSun39.AutomaticFaceReenactment,PabloGarrido,LeviValgaerts,OleRehmsen,ThorstenThormählen,PatrickPérez,ChristianTheobalt
40.JointSummarizationofLarge-scaleCollectionsofWebImagesandVideosforStorylineReconstruction,GunheeKim,LeonidSigal,EricP.Xing
41.Semi-supervisedRelationalTopicModelforWeaklyAnnotatedImageRecognitioninSocialMedia,ZhenxingNiu,GangHua,XinboGao,QiTian
42.BeyondHumanOpinionScores:BlindImageQualityAssessmentbasedonSyntheticScores,PengYe,JayantKumar,DavidDoermann
43.ActiveSamplingforSubjectiveImageQualityAssessment,PengYe,DavidDoermann
44.AStudyonCross-PopulationAgeEstimation,GuodongGuo,ChaoZhang
45.RemoteHeartRateMeasurementFromFaceVideosUnderRealisticSituations,XiaobaiLi,JieChen,GuoyingZhao,MattiPietikäinen
46.6SecondsofSoundandVision:CreativityinMicro-Videos,MiriamRedi,NeilO'Hare,RossanoSchifanella,MicheleTrevisiol,AlejandroJaimes
47.GPS-TagRefinementusingRandomWalkswithanAdaptiveDampingFactor,AmirRoshanZamir,ShervinArdeshir,MubarakShah


Saturday,June28Workshops
ComputationalModelsforSocialInteractions&Behavior:ScientificGrounding,Sensing&Applications
Organizers:AjayDivakaranManeeshSinghMohamedAmerSaadKhanBehjatSiddiquieLocation:C112-113Schedule:FullDay
0910Welcome
S1:InvitedTalks(0915-1025
0915OverviewofAffectiveComputing,RosalindPicard(MIT0945BuildingBlocksofSocialInteraction,BrianLande
(SCPD1015MorningBreakS2:InvitedTalks(1040-1200
1040NeuroscienceofSocialInteractions,WilliamCasebeer
(DARPA1120Neuroscientificaspectsofhumanperception,Ido
Davidesco(PrincetonUniv.1045InvitedTalk:IARPAProgram,MarkBurge1200LunchBreak(ExhibitHallCS3:InvitedTalks(1300-1445
1300AmovementbasedperspectiveonSocialInteractions,
ElizabethTorres(RutgersUniv.1340MultimodalAnalysisofHumanBehavior,Louis-Philippe
Morency(USC1445AfternoonBreak
S4:AfternoonSession(1500-1725
1500InvitedTalk:MachineLearningbasedTemporal
Models,GrahamTaylor(Univ.ofGuelph1540OralPresentations1625PanelDiscussion1725ClosingRemarks
40
Saturday,June28
0700–1700Registration(ExhibitHallCLobby0730–0830Breakfast(ExhibitHallC


Web-scaleVisionandSocialMedia
Organizers:LambertoBallanAlexBergMarcoBertiniThomasMensinkRahulSukthankarLocation:C111Schedule:FullDay
0900Welcome
0910InvitedTalk:Marc’AurelioRanzato(Facebook0950PhotoRecall:UsingtheInternettoLabelYourPhotos,
NeerajKumar,SteveSeitz1015MorningBreak
1040Streetscore-PredictingthePerceivedSafetyofOne
MillionStreetscapes,NikhilNaik,JadePhilipoom,RameshRaskar,CesarHidalgo1100AStreamAlgebraforComputerVisionPipelines,
MohamedHelala,KenPu,FaisalQureshi1120InvitedTalk:Media,Community,andtheSocial
Photograph,DavidAymanShamma(YahooResearch1200LunchBreak(ExhibitHallC
1330InvitedTalk:JulianMcAuley(StanfordUniv.1410Whatisusualinunusualvideos?Trajectorysnippet
histogramsfordiscoveringunusualness,AhmetIscen,AnilArmagan,PinarDuygulu1430ClusteringSocialEventImagesusingKernelCanonical
CorrelationAnalysis,UnaizaAhsan1450PanelDiscussion



Saturday,June28
Long-termDetectionandTracking
Organizers:OctaviaCampsRitaCucchiaraAlbertoDelBimboJiriMatasFedericoPerniciStanSclaroffLocation:C114Schedule:Fullday
0905Opening
0910InvitedTalk:CristophLampert(ISTAustria0945InvitedTalk:PeterMeier(CTOMetaio1015MorningBreak

EmbeddedVision
Workshops
Organizers:GokselDedeogluFridtjofSteinStefanoMattocciaJagadeeshSankaranLocation:C210Schedule:Fullday
0845OpeningRemarks:GokselDedeogluandFridtjofStein,
GeneralChairs0900Keynote:ProjectTango:GivingMobileDevicesa
Human-ScaleUnderstandingofSpaceandMotion,JohnnyLee(Google1000Brain-inspiredClassroomOccupancyMonitoringona
Low-PowerMobilePlatform,FrancescoConti,AntonioPullini,LucaBenini1020FastLBPFaceDetectiononLow-PowerSIMD
Architectures,OlexaBilaniuk,EhsanFazl-Ersi,RobertLaganiere,ChristinaXu,DanielLaroche,CraigMoulder1040InvitedTalk:DeepLearningArchitectures,Eugenio
Culurciello(PurdueUniv.1140LightningTalks:2-minute-per-poster/demobriefs1200LunchBreak(ExhibitHallC1230PostersandDemos
Posters
AHigh-PerformanceHardwareArchitectureforaFramelessStereoVisionAlgorithmImplementedonaFPGAPlatform,FlorianEibensteiner,JürgenKogler,JosefScharinger
TowardsAutonomousNavigationofMiniatureUAV,RolandBrockers,MartinHummenberger,StephanWeiss,LarryMatthies
ATrainStationSurveillanceSystem:ChallengesandSolutions,BurakOzer,MarilynWolf
AddressingSystem-LevelOptimizationwithOpenVXGraphs,ErikRainey,JesseVillarreal,GokselDedeoglu,KariPulli,ThierryLepley,FrankBrill
ACompute-EfficientAlgorithmforRobustEyebrowDetection,SupriyaSathyanarayana,RaviKumar

1045InvitedAbstracts
PersistentPeopleTrackingandFaceCaptureOveraWideArea,GérardMedioni,YinghaoCai
TrackletAssociationinDetect-then-trackParadigmforLong-termMulti-PersonTracking(ExtendedAbstract,BingWang,GangWang,KapLukChan,LiWang
OnFastTrackersthatareRobusttoPartialOcclusions,LuZhang,HamdiDibeklioglu,LaurensvanderMaaten1130DatasetPapers
TheMatrioskaTrackingAlgorithmonLTDT2014Dataset,MarioEdoardoMaresca,AlfredoPetrosino
On-lineVideoMotionEstimationbyInvariantReceptiveInputs,MarcoGori,MarcoLippi,MarcoMaggini,StefanoMelacci1200LunchBreak(ExhibitHallC
1330InvitedTalk:ArnoldSmuelders(Univ.ofAmsterdam1400Paneldiscussion1445Wrap-up
41


Saturday,June28Workshops
LargeScaleVisualRecognitionandRetrieval(BigVision
Organizers:JiaDengAlexBergYuanqingLinJasonCorso
Location:GrandBallroom2Schedule:FullDay
0900OpeningRemark
0905InvitedTalk:TowardsaVisualMemex,AlexeiEfros
(Univ.ofCalifornia,Berkeley0940InvitedTalk:VideoasTrainingDataforObjectClass
Detectors,VittorioFerrari(Univ.ofEdinburgh1015MorningBreak
1045InvitedTalk:TowardsLarge-ScaleSemantic
Representations,TrevorDarrellandYangqingJia(Univ.ofCalifornia,Berkeley1120SpotlightsandPosters1200LunchBreak(ExhibitHallC
1415InvitedTalk:LearningfromDescriptiveText,Tamara
Berg(Univ.ofNorthCarolina1450InvitedTalk:TheDistributedCamera,NoahSnavely
(CornellUniv.1525AfternoonBreak
1555InvitedTalk:Large-ScaleImageUnderstanding,DragoAnguelov(Google1630InvitedTalk:TowardaUniversalPerceptionSystem,
YannLeCun(NewYorkUniv.&Facebook
Satzoda,SuchitraSathyanarayana,SrikanthanThambipillai
AnEmbeddedSolutiontoVisualMappingfor
ConsumerDrones,GuyueZhou,AngLiu,KangYang,TaoWang,ZexiangLi
ASurroundViewCameraSolutionforEmbeddedSystems,BuyueZhang,VikramAppia,IbrahimPekkucuksen,YuchengLiu,AzizUmitBatur,PavanShastry,StanleyLiu,ShijuSivasankaran,KedarChitnisFPGA-basedFastResponseImageAnalysisforAutonomousorSemi-AutonomousIndoorFlight,RobertLadig,KazuhiroShimonomura
ExploitingTrafficSceneDisparityStatisticsforStereoVision,StefanK.Gehrig,UweFranke,NicolaiSchneiderA240G-ops/sMobileCoprocessorforDeepNeuralNetworks,VinayakGokhale,JonghoonJin,AysegulDundar,BerinMartini,EugenioCulurciello
Demos
EfficientLaneandVehicleDetectionwithIntegratedSynergies(ELVIS,RaviKumarSatzoda,MohanM.Trivedi
A240G-ops/sMobileCoprocessorforDeepNeuralNetworks,VinayakGokhale,JonghoonJin,AysegulDundar,BerinMartini,EugenioCulurciello1400KeynoteTalk:EyeTap,SteveMann(Univ.ofToronto1500InvitedTalk:OpenVX,VictorEruhimov(ItSeez1540AfternoonBreak
1600GestureRecognitioninEgo-CentricVideosusingDense
TrajectoriesandHandSegmentation,LorenzoBaraldi,FrancescoPaci,GiuseppeSerra,LucaBenini,RitaCucchiara1620EfficientLaneandVehicleDetectionwithIntegrated
Synergies(ELVIS,RaviKumarSatzoda,MohanM.Trivedi1640InvitedTalk:EmbeddedVisionChallengesfor
ImplementingAugmentedRealityApplications,PeterMeier(MetaIO1720ClosingRemarks
42





Saturday,June28
EgocentricVision
Organizers:KrisKitaniYongJaeLeeMichaelS.RyooAlirezaFathi
Location:GrandBallroom3Schedule:FullDay

Workshops
1400KeynoteTalk:Whois"Ego"inEgo-centricVision?,Ben
Kuipers(Univ.ofMichiganSpotlightsandPosterSessionB(1445-1645
FromEgotoNos-vision:DetectingSocialRelationshipsinFirst-PersonViews,StefanoAlletto,GiuseppeSerra,SimoneCalderara,FrancescoSolera,RitaCucchiaraASequentialClassifierforHandDetectioninthe
FrameworkofEgocentricVision,AlejandroBetancourt,MiriamM.López,CarloS.Regazzoni,MatthiasRauterbergEye-Model-BasedGazeEstimationbyRGB-DCamera,LiJianfeng,LiShigang
ExperimentsonanRGB-DWearableVisionSystemforEgocentricActivityRecognition,MohammadMoghimi,PabloAzagra,LuisMontesano,AnaC.Murillo,SergeBelongie
ParsingVideosofActionswithSegmentalGrammars,HamedPirsiavash,DevaRamanan
EstimatingRelativeSocialStatusfromFace-to-FaceInteractionsusingFirst-personVision,MiraiHiguchi,KrisM.Kitani,YoichiSato
WearableRGB-DNavigationSystemForTheBlind,YoungHoonLee,GérardMedioni
VideoSET:VideoSummaryEvaluationToolkit,SerenaYeung,AlirezaFathi,LiFei-Fei
First-PersonActivityRecognitionfromAnimalVideos,YumiIwashita,AsamichiTakamine,RyoKurazume,MichaelS.Ryoo
2DHandParsingforEgocentricGestureRecognition,AkankshaSaran,KrisM.Kitani
ObjectRecognitioninEgocentricVideoswithSaliency-basedNon-uniformSamplingandVariableResolutionSpaceforFeaturesSelection,VincentBuso,JennyBenois-Pineau,IvánGonzález-Díaz
GazeEstimationusingFingertipGazeCalibration,TakeshiSaitoh
IndoorTrajectoryEstimationfromWearableCameraforActivityMonitoring,GuillaumeBourmaud,RémiMégret,AudreyGiremus,YannickBerthoumieu
SummarizationofEgocentricMovingVideosfor
GeneratingWalkingRouteGuidance,MasayaOkamoto,YoshiyukiKawano,KeijiYanai
0930KeynoteTalk:TakeoKanade(CMU0945KeynoteTalk:ChiekoAsakawa(IBM
SpotlightsandPosterSessionA(1030-1200
ActionandInteractionRecognitioninFirst-Person
Videos,SanathNarayan,MohanS.Kankanhalli,KalpathiR.Ramakrishnan
Video-basedObjectRecognitionUsingNovelSet-of-SetsRepresentations,YangLiu,YoungkyoonJang,WoontackWoo,Tae-KyunKim
EfficientRetrievalfromLarge-ScaleEgocentricVisualDataUsingaSparseGraphRepresentation,Vijay
Chandrasekhar,WuMin,XiaoLi,ChestonTan,BappadityaMandal,LiyuanLi,JooHweeLim
UnderstandingtheNatureofFirst-PersonVideos:CharacterizationandClassificationusingLow-LevelFeatures,ChestonTan,HanlinGoh,VijayChandrasekhar,LiyuanLi,Joo-HweeLim
ThisHandIsMyHand:AProbabilisticApproachtoHandDisambiguationinEgocentricVideo,StefanLee,SvenBambach,DavidJ.Crandall,JohnM.Franchak,ChenYuAnAttention-basedActivityRecognitionforEgocentricVideo,KenjiMatsuo,KentaroYamada,SatoshiUeno,SeiNaito
Temporally-DependentDirichletProcessMixturesforEgocentricVideoSegmentation,JosephW.Barker,JamesW.Davis
VisualNavigationAidfortheBlindinDynamicEnvironments,Tung-SingLeung,GérardMedioni
WisdomoftheCrowdinEgocentricVideoCuration,YedidHoshen,GilBen-Artzi,ShmuelPeleg1200LunchBreak(ExhibitHallC
1315KeynoteTalk:AMeasureandTheoryof3DJoint
AttentionfromFirstPersonCameras,YaserShiekh(CMU
43


Saturday,June28Workshops
Multi-SensorFusionforOutdoorDynamicSceneUnderstanding
Organizers:MubarakShahWolfgangFörstnerAlperYilmazClémentMalletMichaelYingYangYuryVizilterLocation:C220-221
Schedule:HalfDay—Morning
0830Welcome
0840IntegratingLIDARRangeScansandPhotographswith
TemporalChanges,BrittanyMorago,GiangBui,YeDuan0900GuidedDepthUpsamplingviaACosparseAnalysis
Model,XiaojinGong,JianqiangRen,BaishengLai,ChaohuaYan,HuiQian0920KeynoteTalk:RaquelUrtasun(Univ.ofToronto1015MorningBreak1045PosterSpotlights
1120PosterSession
Alignmentof3DBuildingModelswithSatelliteImagesUsingExtendedChamferMatching,XiZhang,GadyAgam,XinChen
ActivePlanning,SensingandRecognitionUsinga
Resource-ConstrainedDiscriminantPOMDP,ZhaowenWang,ZhangyangWang,MarkMoll,Po-SenHuang,DevinGrady,NasserNasrabadi,ThomasHuang,LydiaKavraki,MarkHasegawa-Johnson
FrameRateFusionandUpsamplingofEO/LIDARDataforMultiplePlatforms,T.NathanMundhenk,KyungnamKim,YuriOwechko
FeatureRegressionforMultimodalImageAnalysis,MichaelYingYang,XuanziYong,BodoRosenhahn2D/3DSensorExploitationandFusionforEnhancedObjectDetection,JiejunXu,KyungnamKim,ZhiqiZhang,Hai-wenChen,YuriOwechko

3DHandPoseDetectioninEgocentricRGB-DImages,GrégoryRogez,MaryamKhademi,JamesStevenSupančičIII,J.M.M.Montiel,DevaRamanan
3-DGazeScanPathbyInside-outCameraSystem,
HironobuFujiyoshi,MakotoKimura,ShoichiShimizu,YujiYamauchi,TakayoshiYamashita
PlaceAvoider:SteeringFirst-PersonCamerasAwayfromSensitiveSpaces,RobertTempleman,MohammedKorayem,DavidCrandall,ApuKapadia1645ClosingRemarksandAwardPresentations
ComputationalCameras&Displays
Organizers:AshokVeeraraghavanOliverCossairtKaushikMitraLocation:C215-216
Schedule:HalfDay—Morning
0830Welcome
0845InvitedTalk:IntegratedImaging:CreatingImages
fromtheTightIntegrationofAlgorithms,
Computation,andSensors,CharlesA.Bouman(PurdueUniv.0930ANovelHDRDepthCameraforReal-time3D360°
PanoramicVision,AhmedNabilBelbachir,StephanSchraml,ManfredMayerhofer,MichaelHofstätter0945SeparatingTextureandIlluminationforSingle-Shot
StructuredLightReconstruction,MinhVo,SrinivasaG.Narasimhan,YaserSheikh1000MorningBreak:Posters&DemosSession
1045InvitedTalk:TimeofFlightRevolution,MohitGupta
(ColumbiaUniv.1130LightFieldScale-DepthSpaceTransformforDense
DepthEstimation,IvanaTošić,KathrinBerkner1145ProjectionCenterCalibrationforaCo-located
ProjectorCameraSystem,ToshiyukiAmano1200DictionaryLearningbasedColorDemosaicingfor
PlenopticCameras,XiangHuang,OliverCossairt1215BestPaperAward&ConcludingRemarks
44



Saturday,June28
ChangeDetection
Organizers:Pierre-MarcJodoinJanuszKonradPrakashIshwarFatihPorikliLocation:C220-221
Schedule:HalfDay—Afternoon
1300OpeningRemarks&DescriptionoftheChallenge
Workshops
DeepVision:DeepLearninginComputerVision
Organizers:JoseM.AlvarezYannLeCunFatihPorikliYiLi
Location:GrandBallroom1Schedule:HalfDay—Afternoon
1300OpeningRemarks1305InvitedTalk:KayYu(Baidu
1400HeterogeneousMulti-taskLearningforHumanPose
EstimationwithDeepConvolutionalNeuralNetwork,SijinLI,Zhi-QiangLiu,AntoniB.Chan1415GeneralizedAutoencoder:ANeuralNetwork
FrameworkforDimensionalityReduction,WeiWang,YanHuang,YizhouWang,LiangWang1430UnrollingLoopyTop-downSemanticFeedbackin
ConvolutionalDeepNetworks,CarloGatta,AdrianaRomero,JoostvandeVeijer1445CNNFeaturesOff-the-Shelf:AnAstoundingBaseline
forRecognition,AliSharifRazavian,HosseinAzizpour,JosephineSullivan,StefanCarlsson1500APiggybackRepresentationforActionRecognition,
LiorWolf,YairHanani,TalHassner1515PosterSession(withAfternoonBreak1555Awards
1600InvitedTalk:PierreSermanet(NewYorkUniv.1650ConcludingRemarks
1325AFastSelf-TuningBackgroundSubtractionAlgorithm,
BinWang,PiotrDudek1350Spectral-360:APhysics-BasedTechniqueforChange
Detection,MohamedSedky,MansourMoniri,ClaudeC.Chibelushi1415Break
1430ChangeDetectionwithWeightlessNeuralNetworks,
MassimoDeGregorio,MaurizioGiordano1455FlexibleBackgroundSubtractionWithSelf-Balanced
LocalSensitivity,Pierre-LucSt-Charles,Guillaume-AlexandreBilodeau,RobertBergevin1520AfternoonBreak
1550StaticandMovingObjectDetectionUsingFluxTensor
withSplitGaussianModels,RuiWang,FilizBunyak,GunaSeetharaman,KannappanPalaniappan1615Conclusion&FutureWork


45


Saturday,June28
TowardsSolvingReal-worldVisionProblemswithRGB-DCameras
Organizer:JuergenGallXiaofengRenPushmeetKohliTime:Location:
0830-1700(FullDayC211-212

Organizer:ZaidHarchaouiHervéJégouFlorentPerronninTime:Location:
0830-1700(FullDayC215
Tutorials
Large-ScaleVisualRecognition
Description:RGB-Ddepthcamerashavethepotentialtobecomeakeycomponentforsolvingreal-worldproblems.Theirlowcostandwidespreadavailabilityhavemadethemacommercialsuccessandtheirpopularityintheresearchcommunityhasdramaticallyincreased.Inthemeanwhile,thenextgenerationofRGB-Dsensorshavebeendevelopedthatarebetterthancurrentsensorsintermsofdepthquality,frame-rate,andsensorsize.Itisthereforeagoodmomenttosummarizewhathasbeenachievedsofarandtodiscussinterestingpossibledirectionforfuturework.Theproposedshortcourseintendstodiscussthebasics,underlyingprinciplesandcutting-edgeresultsofacomprehensivelistoftopicsinRGB-Dperception.
Description:ThistutorialaddressesthetopicofLarge-ScaleVisualRecognition(LSVR,theproblemofunderstandingvisualcontent(e.g.photosorvideosonalarge-scale.Thistopichasreceivedmuchattentioninthecomputervisioncommunityinthelastfewyears:aslargerdatasetshavebecomeavailable,handlingmillionsofimagesandthousandsoflabelclasseshasbecomethenormratherthantheexception.SinceLSVRisabroadtopic,wewillmainlyfocusontwotasks:imageretrievalandimageclassification.Thegoalsofthistutorialarefour-fold:
Providetheaudiencewiththekey"tools"toprocesssuchlargedatasets.Reviewthemainfamiliesoffeaturesforvisualrecognition,frombag-of-visual-wordstodeepconvolutionalfeatures,includingVLADandFisherVectorsShowtheconvergencebetweenlarge-scaleretrievalandlarge-scaleclassification,twoproblemswhichhavebeentraditionallyaddressedseparately.ShowthatLSVRdoesnotnecessarilyrequiremassivecomputationalresources(althoughsuchresourcescanhelp,ofcourse...

VisualSLAM
Organizer:FrankDellaertMichaelKaessTime:Location:
0830-1700(FullDayC214
Description:ThistutorialaddressesVisualSLAM,theproblemofbuildingasparseordense3Dmodelofthescenewhiletravelingthroughit,andsimultaneouslyrecoveringthetrajectoryoftheplatform/camera.VisualSLAMhasreceivedmuchattentioninthecomputervisioncommunityinthelastfewyears,asmorechallengingdatasetsbecomeavailable,andvisualSLAMisstartingtobeimplementedonmobilecamerasandusedinARandotherapplications.Wewillprovideanintroductiontothecoreconceptsunderlyingcurrentsparse,denseandsemanticvisualSLAMsystems.
46



Saturday,June28
DescribingImagesinNaturalLanguage
Organizer:JuliaHockenmaierTime:Location:
0830-1230(HalfDay—MorningC115

Organizer:TillQuackJanErikSolemTime:Location:
Tutorials
FoundingaComputerVisionStartup
0830-1230(HalfDay—MorningC213
Description:Theabilitytoassociateimageswithnaturallanguagesentencesthatdescribewhatisdepictedinthemisahallmarkofimageunderstanding,andaprerequisiteforapplicationssuchassentence-basedimagesearch.Thepurposeofthistutorialistogiveresearchersincomputervisionanoverviewoftheissuesinvolvedinautomaticimagedescription,andtointroducethemtonaturallanguageprocessingtoolsandideastheycanuseforthispurpose.

LearningandInferenceinDiscreteGraphicalModels
Organizer:
Time:Location:
NikosKomodakisNikosParagiosDhruvBatraStephenGould
1300-1700(HalfDay—AfternoonC115
Description:AfewyearsagoJanErikandTillbothfoundedstartupsinthecomputervisionfield:PolarRoseandKooaba,respectively.Bothcompanieshadtheirsuccessfulexitswithaglobalplayer,andJanErikevenwentontofoundhisnextcompany.Alongthe-sometimesbumpy-journeywelearnedagreatdealofthingswhichwewouldliketosharewiththecomputervisioncommunity.WefoundthatcreatingyourownstartupisaveryfulfillingexperienceandaninterestingcareerchoiceafteracompletedPhD.Inaddition,thetimeisrighttocreatemorestartupsinthevisionfield,wheretechnologyisnowinacyclewheremanynovel(consumerapplicationsbecomefeasible.
WegaveaversionofthiscourseatCVPR2010.Muchhaschangedsince,butthegoalofthiscourseremainstogivearatherbroadoverviewofpracticesandtoolswhichturnedouttobeusefulforus.Nowrefreshedandimprovedwithadditionallearnings.
ThecoursespansdiversetopicsasFunding,Hiring,SoftwareEngineering,BusinessModels,ProductDesign.Wepresentmaterialforeachtopic,togetherwithpointersandlinkstorelevantresources.InthisneweditionofthetutorialwewillalsofocusonconversationsbetweenJanErikandTillonthedifferenttopics.
Description:Severalproblemsincomputervisioncanbeformulatedusingthediscretegraphicalmodelsframework.Thetwomainissuesfacedbyresearchersinthiscaseare:(iLearning:Howtoestimatetheparametersofthemodel?;and(iiInference:Howtofindthebestassignmentforthevariablesofthemodel?Inthistutorialwewilldiscussthesetwoissues,startingfromthebasicsandbuildinguptothestateoftheart.
47


CVPR2014Notes
48


本文来源：https://www.2haoxitong.net/k/doc/db79df9e90c69ec3d4bb755a.html

《CVPR_2014_Pocket_Guide_Final.doc》
将本文的Word文档下载到电脑，方便收藏和打印
推荐度：
点击下载文档

文档为doc格式
相
关
案
例

相关推荐

推荐内容

【2018最新】联合国秘书长潘基文在国际埃博拉恢复会议英语演讲稿-实用word文档 (5页) 2015年贵州财大会计专硕(mpacc)难易程度分析 2013年全国数据基础理论要领 教资面试高中英语试讲模板之阅读听力 Lending Club 如何修炼成为全球最大P2P平台 励志语录大全 美元指数简介 2011年店长工作总结 茶文化精髓所在 (500-700kg护面石安放)验收、评定表
网站内容来自网络，如有侵权请联系我们，立即删除！ Copyright © 范文写作网京ICP备16605803号