王金桥男博导自动化研究所 一直从事视频分析与检索、目标检测与跟踪、图像分类与识别方面的应用基础研究，同时也涉及了模式识别与机器学习方面的相关理论研究，围绕大规模视觉识别和视频大数据分析成立了智能媒体计算联合实验和视频大数据云识别联合实验。共发表论200余篇，国际杂志36篇，国际会议160余篇。其中以第一作者或通讯作者发表高水平国际期刊论文30篇，包括多媒体权威期刊（IEEETrans.onMultimedia）、图像处理权威期刊（IEEETrans.onImageProcessing）、神经网络与学习系统会刊（IEEETransactionsonNeuralNetworksandLearningSystem）、视频处理权威期刊（IEEETrans.CircuitsandSystemsforVideoTechnology）、模式识别（PatternRecognition）；其中发表重要国际会议论文40余篇，包括多媒体顶级会议（ACMMultimedia），计算机视觉顶级会议（ICCV），模式识别与计算机视觉顶级会议（CVPR）和人工智能顶级会议（AAAI，IJCAI）等。

招生专业

081104-模式识别与智能系统

招生方向

视频大数据分析与检索 目标检测与跟踪

模式识别与机器学习 教育背景

2004-09--2008-07中科院自动化研究所博士 2001-09--2004-07天津大学硕士 工作简历

2016-10~现在,中科院自动化所,研究员 2011-10~2016-09,中科院自动化研究所,副研究员 2008-07~2011-10,中科院自动化研究所,助理研究员

国内外竞赛 2011年微软亚洲研究院“铸星计划”青年访问学者。 2014年第一届智慧城市视频分析技术挑战赛中，获得行人检测第二名、多目标检测第二名。 2015年第二届全国研究生智慧城市技术与创意设计大赛中视频挑战赛的5个竞赛任务中，获得第一名2项、第二名3项（参赛队伍117支）； 2015年欧洲图形国际会议多视角目标检索竞赛第一名，其中6项指标4项第一。 2016年由美国国家标准与技术研究院NIST组织TRECVID视频实例检索（InstanceRetrieval）比赛第一名。 2016上海BOT（BRAINOFTHINGS万物之智）视觉大数据识别竞赛第一名和最佳算法奖（全球400队），奖金20万元。 2016CCF大数据与计算智能大赛监控场景下人的精细化识别二等奖

专利成果

（1）基于动态群组划分的多目标跟踪算法，发明，2011，第2作者，专利号：201110113290.7 （2）极小化上界误差的在线协同提升学习的视觉跟踪方法，发明，2010，第2作者，专利号：201010603293.1 （3）基于目标分类和轨迹聚类的语义场景模型学习方法，发明，2010，第2作者，专利号：201010603240.2 （4）基于群组上下文的行人计数方法，发明，2011，第2作者，专利号：201110201983.1 （5）商标检测与识别方法，发明，2011，第2作者，专利号：PCT/CN2011/073985 （6）基于视频监控网络的自动视频浓缩技术，发明，2011，第3作者，专利号：201110208090.X

科研项目

（1）面向移动终端的视频检索与浏览，主持，国家级，2013-01--2016-12 （2）基于语义分析和视觉关注的视频自适应研究，主持，国家级，2010-01--2012-12 （3）数字媒体理解的理论与方法研究，参与，国家级，2010-01--2015-12

图像与视频分析大规模目标检测与识别目标跟踪目标检索

担任InternationalConferenceonInternetMultimediaComputingandService（ICIMCS2009）出版主席。担任ICME2011和MMM2011、ICIP2014和ACMMultimedia2015的程序委员会委员。担任IEEETrans.onMultimedia、ComputerVisionandImageUnderstanding、MultimediaSystems、IEEETrans.CircuitsandSystemsforVideoTechnology、Neurocomputing等国际期刊杂志以及CVPR、ICME、ICIP、ICPR、ACCV、MMM、PCM、ACMMultimedia等国际会议的审稿人。 ICIMCS2009（互联网多媒体计算与服务国际会议）的出版主席; PCM2011和MMM2011的程序委员会委员;

2016

[1]RenYang,HuazhongXu,JinqiaoWang:RobustCrowdSegmentationandCountinginIndoorScenes.MMM(1)2016:505-514 [2]GuiboZhu,JinqiaoWang,YiWu,XiaoyuZhang,HanqingLu:MC-HOGCorrelationTrackingwithSaliencyProposal.AAAI2016 [3]Real-timeVisualTrackingwithMultipleMulti-kernelCorrelationFilters,Underreview,(Thesourcecodewillbeavailableafterthepaperisaccepted).[Video-Results] [4]ChaoyangZhao,JinqiaoWang,GuiboZhu,YiWu,HanqingLu:Learningweightedpartmodelsforobjecttracking.ComputerVisionandImageUnderstanding143:173-182(2016) [5]JinqiaoWang,MinXu,HanqingLu,IanBurnett:ActiveAd:Anovelframeworkoflinkingadvideostoonlineproducts.Neurocomputing185:82-92(2016) [6]JunLuo,JinqiaoWang,HuazhongXu,HanqingLu:Real-timepeoplecountingforindoorscenes.SignalProcessing124:27-35(2016) [7]YingyingChen,JinqiaoWang,MinXu,XiangjianHe,HanqingLu:Aunifiedmodelsharingframeworkformovingobjectdetection.SignalProcessing124:72-80(2016) [8]HuiyingLiu,MinXu,JinqiaoWang,TianrongRao,IanBurnett:ImprovingVisualSaliencyComputingWithEmotionIntensity.IEEETrans.NeuralNetw.LearningSyst.27(6):1201-1213(2016) [9]ZhengWang,RuiminHu,YiYu,JunjunJiang,ChaoLiang,JinqiaoWang:Scale-AdaptiveLow-ResolutionPersonRe-IdentificationviaLearningaDiscriminatingSurface.IJCAI2016:2669-2675 [10]GuiboZhu,JinqiaoWang,HanqingLu:Clusteringbasedensemblecorrelationtracking.ComputerVisionandImageUnderstanding153:55-63(2016) [11]WeijianRuan,JunChen,JinqiaoWang,BoLuo,WenjunHuang,RuiminHu:Boostedlocalclassifiersforvisualtracking.ICME2016:1-6 [12]JinqiaoWang,ZhanQu,YingyingChen,TaoMei,MinXu,LaZhang,HanqingLu:AdaptiveContentCondensationBasedonGridOptimizationforThumbnailImageGeneration.IEEETrans.CircuitsSyst.VideoTechn.26(11):2079-2092(2016) [13]HaiyunGuo,JinqiaoWang,YueGao,JianqiangLi,HanqingLu:Multi-View3DObjectRetrievalWithDeepEmbeddingNetwork.IEEETrans.ImageProcessing25(12):5526-5537(2016) [14]ZhiqiangLi,LaZhang,YikaiFang,JinqiaoWang,HuazhongXu,BaocaiYin,HanqingLu:DeepPeopleCountingwithFasterR-CNNandCorrelationTracking.ICIMCS2016:57-60 [15]GuanzhongWang,YikaiFang,JinqiaoWang,JianSun:ExtensiveComparisonofVisualFeaturesforPersonRe-identification.ICIMCS2016:192-196 [16]TianrongRao,MinXu,HuiyingLiu,JinqiaoWang,IanBurnett:Multi-scaleblocksbasedimageemotionclassificationusingmultipleinstancelearning.ICIP2016:634-638 [17]LingxiangWu,JinqiaoWang,GuiboZhu,MinXu,HanqingLu:Personre-identificationviarichcolor-gradientfeature.ICME2016:1-6

2015

[1]JunLuo,JinqiaoWang,HuazhongXu,HanqingLu:AReal-TimePeopleCountingApproachinIndoorEnvironment,MMM2015 [2]GuiboZhu,JinqiaoWang,YiWu,HanqingLu:CollobarativeCorrelationTracking,BMVC2015[PDF][PPT][CODE] [3]JinqiaoWang,JianlongFu,HanqingLu:Findinglogosinreal-worldimageswithpoint-contextrepresentation-basedregionsearch.MultimediaSyst.21(3):301-311(2015) [4]JianlongFu,JinqiaoWang,YongRui,Xin-JingWang,TaoMei,HanqingLu:ImageTagRefinementWithView-DependentConceptRepresentations.IEEETrans.CircuitsSyst.VideoTechn.25(8):1409-1422(2015) [5]Guibozhu,JinqiaoWang,ChaoyangZhao,HanqingLu:WeightedPartContextLearningforVisualTracking.IEEETransactionsonImageProcessing24(12):5140-5151(2015) [6]HaiyunGuo,JinqiaoWang,YingyingChen,JianlongFu,ChaoyangZhao,GuiboZhuandHanqingLu:3DObjectRetrievalwithMultimodalViews.3DOR2015:129-136 [7]WeiFu,ChaoyangZhao,JinqiaoWang,JingLiu,JianCheng,HanqingLu:Concurrentgroupactivityclassificationwithcontextmodeling.ICIMCS2015:9 [8]JunLuo,JinqiaoWang,LaZhang,YingyingChen,HuazhongXu,HanqingLu:60Hzself-tuningbackgroundmodeling.ICIMCS2015:78 [9]YingyingChen,JinqiaoWang,HanqingLu:Learningsharablemodelsforrobustbackgroundsubtraction.ICME2015:1-6 [10]YingyingChen,JinqiaoWang,JingLiu,HanqingLu:MobileMediaThumbnailing.ICMR2015:665-666 [11]HaiyunGuo,JinqiaoWang,MinXu,Zheng-JunZha,HanqingLu:LearningMulti-viewDeepFeaturesforSmallObjectRetrievalinSurveillanceScenarios.ACMMultimedia2015:859-862 [12]YuhangWang,JingLiu,JinqiaoWang,YongLi,HanqingLu:Colornameslearningusingconvolutionalneuralnetworks.ICIP2015:217-221 [13]HaiyunGuo,JinqiaoWang,HanqingLu:Learningdeepcompactdescriptorwithbaggingauto-encodersforobjectretrieval.ICIP2015:3175-3179 [14]YingyingChen,JinqiaoWang,JianqiangLi,HanqingLu:Multiplefeaturesbasedsharedmodelsforbackgroundsubtraction.ICIP2015:3946-3950 [15]JianlongFu,YueWu,TaoMei,JinqiaoWang,HanqingLu,YongRui:RelaxingFromVocabulary:RobustWeakly-SupervisedDeepLearningforVocabulary-FreeImageTagging.ICCV2015 [16]GuiboZhu,JinqiaoWang,ChaoyangZhao,HanqingLu:WeightedPartContextLearningforVisualTracking.IEEETransactionsonImageProcessing24(12):5140-5151(2015)

2014

[1]WeiFu,JinqiaoWang,LiangkeGui,HanqingLu,SongdeMa:Onlinevideosynopsisofstructuredmotion.Neurocomputing135:155-162(2014) [2]LifangWu,LianchaoCao,MinXu,JinqiaoWang:AHybridImageRetargetingApproachviaCombiningSeamCarvingandGridWarping.JournalofMultimedia9(4):483-492(2014) [3]MinXu,JinqiaoWang,XiangjianHe,JesseS.Jin,SuhuaiLuo,HanqingLu:Athree-levelframeworkforaffectivecontentanalysisanditscasestudies.MultimediaToolsAppl.70(2):757-779(2014) [4]JinqiaoWang,BoWang,Ling-YuDuan,QiTian,HanqingLu:Interactiveadsrecommendationwithcontextualsearchonproducttopicspace.MultimediaToolsAppl.70(2):799-820(2014) [5]XiaobinZhu,JingLiu,JinqiaoWang,HanqingLu:Keyobservationselection-basedeffectivevideosynopsisforcameranetwork.Mach.Vis.Appl.25(1):145-157(2014) [6]XiaobinZhu,JingLiu,JinqiaoWang,ChangshengLi,HanqingLu:Sparserepresentationforrobustabnormalitydetectionincrowdedscenes.PatternRecognition47(5):1791-1799(2014) [7]JinqiaoWang,MinXu,XiangjianHe,HanqingLu,DoanHoang:Ahybriddomainenhancedframeworkforvideoretargetingwithspatial-temporalimportanceand3Dgridoptimization.SignalProcessing94:33-47(2014) [8]BingLi,Ling-YuDuan,JinqiaoWang,RongrongJi,Chia-WenLin,WenGao:SpatiotemporalGridFlowforVideoRetargeting.IEEETransactionsonImageProcessing23(4):1615-1628(2014) [9]JianCheng,TingYuan,JinqiaoWang,HanqingLu:Grouplatentfactormodelforrecommendationwithmultipleuserbehaviors.SIGIR2014:995-998 [10]GuiboZhu,JinqiaoWang,ChaoyangZhao,HanqingLu:PartContextLearningforVisualTracking,BMVC2014,(oral,7.7%) [11]ChaoyangZhao,WeiFu,JinqiaoWang,XiaoBai,QingshanLiu,HanqingLu:DiscriminativeContextModelsforCollectiveActivityRecognition,ICPR2014 [12]YuchenJing,XiuzhenZhang,LifangWu,JinqiaoWang,ZemengFeng,DanWang:RecommendationforCommunitiesonFlickrCombiningUserRatingsandItemImportance,ICME2014 [13]ChaoyangZhao,JinqiaoWang,ChengliXie,HanqingLu:ACoarse-To-FineLogoRecognitionMethodInVideoStreams,ICME2014 [14]YuanGao,LifangWu,YuxinMao,JinqiaoWang:ACurvatureFilterandNormalClusteringBasedApproachtoDetectingCylinderon3DMedicalModel.PCM2014:221-226 [15]JinqiaoWang,WeiFu,JingjingLiu,HanqingLu:SpatiotemporalGroupContextforPedestrianCounting.IEEETrans.CircuitsSyst.VideoTechn.24(9):1620-1630,2014 [16]GuiboZhu,JinqiaoWang,HanqingLu:ClusteringEnsembleTracking,ACCV2014 [17]BingyuanLiu,JingLiu,JinqiaoWang,HanqingLu:LearningaRepresentativeandDiscriminativePartModelwithDeepConvolutionalFeaturesforSceneRecognition,ACCV2014 [18]JianlongFu,JinqiaoWang,TaoMei,YongRui,HanqingLu:WhatVisualAttributesCharacterizeanObjectClass?,ACCV2014 [19]KezhenTeng,JinqiaoWang,HanqingLu:MaskAssistedObjectCodingwithDeepLearningforObjectRetrievalinSurveillanceVideos,ACMMultimedia2014 [20]HuiyingLiu,MinXu,XiangjianHe,JinqiaoWang:EstimateGazeDensitybyIncorporatingEmotion,ACMMultimedia2014 [21]GuiboZhu,JinqiaoWang,HanqingLu:ObjectTrackingwithPart-BasedDiscriminativeContextModels,ICIP2014 [22]HaiyunGuo,KezhenTeng,JinqiaoWang:VideoCopyRetrievalBasedonLocalContextDescriptor,HHME2014 2013

[1]WeiFu,JinqiaoWang,HanqingLu,SongdeMa:Dynamicsceneunderstandingbyimprovedsparsetopicalcoding.PatternRecognition46(7):1841-1850(2013) [2]ZhanQu,JinqiaoWang,MinXu,HanqingLu:Context-AwareVideoRetargetingviaGraphModel.IEEETransactionsonMultimedia15(7):1677-1687(2013) [3]BoWang,JinqiaoWang,HanqingLu:ExploitingcontentrelevanceandsocialrelevanceforpersonalizedadrecommendationoninternetTV.TOMCCAP9(4):26(2013) [4]XianWang,JinqiaoWang:BrandImageDetectioninBroadcastVideoStreams.ICIG2013:852-856 [5]KezhenTeng,JinqiaoWang:ClassificationRelatedManifoldDimensionEstimationwithRestrictedBoltzmannMachine.ICIG2013:857-862 [6]KezhenTeng,JinqiaoWang,QiTian,HanqingLu:Improvingsceneclassificationwithweaklyspatialsymmetryinformation.ICIP2013:3259-3263 [7]BiaoNiu,YifanZhang,JinqiaoWang,JianCheng,HanqingLu:Subspacelearningbasedactivelearningforimageretrieval.ICMEWorkshops2013:1-4 [8]GuiboZhu,JinqiaoWang,ChangshengLi,HanqingLu:CollaborativeTracking:DynamicallyFusingShort-TermTrackersandLong-TermDetector.MMM(2)2013:457-467 [9]ChengyueSun,QiZou,WeiFu,JinqiaoWang:MultipleHypothesesBasedSpatial-TemporalAssociationforStablePedestrianCounting.PCM2013:803-810

2012

[1]YiWu,JianCheng,JinqiaoWang,HanqingLu,JunWang,HaibinLing,ErikBlasch,LiBai.Real-TimeProbabilisticCovarianceTrackingWithEfficientModelUpdate.IEEETransactionsonImageProcessing(TIP),21(5),pp2824-2837,2012. [2]XianXiao,ChangshengXu,JinqiaoWangandMinXu.Enhanced3DModelingforLandmarkImageClassification,IEEETransactiononMultimedia(TMM),2012. [3]WeiFu,JinqiaoWang,ChaoyangZhao,HanqingLuandSongdema,Object-centeredNarrativeforSurveillanceVideo,InternationalConferenceonImageandProcessing(ICIP),2012,pp29-32. [4]ZhanQu,JinqiaoWang,MinXu,HanqingLu,AGridBasedResizingFrameworkviaCombiningWarpingandCropping.InternationalConferenceonImageandProcessing(ICIP),2012,pp2997-3000. [5]XiaobinZhu,JingLiu,JinqiaoWang,YikaiFang,HanqingLu,AbnormalDetectioninCrowdSceneViaAppearanceandDynamicsJointModeling.InternationalConferenceonImageandProcessing(ICIP),2012,pp2525-2531. [6]WeiFu,JinqiaoWang,ZechaoLi,HanqingLuandSongdeMa,LearningSemanticMotionPatternsforDynamicScenesbyImprovedSparseTopicalCoding,IEEEInternationalConferenceonMultimediaandExpo(ICME),2012,pp296-301. [7]ZiYe,JinqiaoWang,ZhenchongWang,HanqingLu:Multiplefeaturesfusionforcrowddensityestimation.InternationalConferenceonInternetMultimediaComputingandService(ICIMCS),2012,pp42-45. [8]YuPeng,SuhuaiLuo,MinXu,ZefengNi,JesseS.Jin,JinqiaoWang,GangZhao:Bagoffeaturesusingsparsecodingforgenderclassification.InternationalConferenceonInternetMultimediaComputingandService(ICIMCS),2012,pp80-83 [9]LianchaoCao,LifangWu,JinqiaoWang:Fastseamcarvingwithstripconstraints.InternationalConferenceonInternetMultimediaComputingandService(ICIMCS),2012,pp148-152. [10]JianlongFu,JinqiaoWang,YifanZhang,HanqingLu:Point-contextdescriptorbasedregionsearchforlogorecognition.InternationalConferenceonInternetMultimediaComputingandService(ICIMCS),2012,pp188-191. [11]ChengliXie,JinqiaoWang,YifanZhang,HanqingLu:Real-timemultipleobjectinstancesdetection.ACMMultimedia2012,pp1301-1302. [12]JianlongFu,JinqiaoWang,ZechaoLi,MinXu,HanqingLu,EfficientClothingRetrievalwithSemantic-PreservingVisualPhrases,AsianConferenceonComputerVision(ACCV),2012,PT1-14. [13]ZhanQu,JinqiaoWang,MinXu,HanqingLu,FusingWarping,Cropping,andScalingforOptimalImageThumbnailGeneration,AsianConferenceonComputerVision(ACCV),2012,pp. [14]XiaobinZhu,JingLiu,JinqiaoWang,WeiFu,HanqingLu,WeightedInteractionForceEstimationforAbnormalityDetectioninCrowdScenes,AsianConferenceonComputerVision(ACCV),2012,pp.

2011

[1]ShiChen,JinqiaoWang,YiOuyang,BoWang,ChangshengXu,HanqingLu,Boostinglearnerstowardeffectiveobjectdetection,ComputerVisionandImageUnderstanding,Vol,115(3),pp364-374,2011. [2]Ling-yuDuan,JinqiaoWang,WenGao,HanqingLu,JesseS.JinandChangshengXu,LinkingTraditionalTVAdvertisingtoInternetAdvertising,IGIGlobalPress2011.(BOOKChapter) [3]YangZhang,JinqiaoWang,WeiFu,HanqingLu,HuazhongXu:Specificvehicledetectionandtrackinginroadenvironment.InternationalConferenceonInternetMultimediaComputingandService(ICIMCS),pp182-186,2011. [4]MinXu,JinqiaoWang,MuhammadAbulHasan,XiangjianHe,ChangshengXu,HanqingLu,JesseS.Jin:Usingcontextsaliencyformovieshotclassification.InternationalConferenceonImageProcessing(ICIP),pp3653-3656,2011. [5]BingLi,YimingChen,JinqiaoWang,Ling-YuDuan,WenGao:Fastretargetingwithadaptivegridoptimization.IEEEInternationalConferenceonMultimediaandExpo(ICME),pp1-4,2011. [6]JingjingLiu,JinqiaoWang,HanqingLu:AdaptiveModelforRobustPedestrianCounting.InternationalConferenceonMultimediaModeling(MMM),pp481-491,2011. [7]BingLi,Ling-YuDuan,JinqiaoWang,JieChen,RongrongJi,WenGao:Grid-BasedRetargetingwithTransformationConsistencySmoothing.InternationalConferenceonMultimediaModeling(MMM),pp12-24,2011. [8]ChengliXie,JianguoLi,TaoWang,JinqiaoWang,HanqingLu,Real-timeCascadeTemplateMatchingforObjectInstanceDetection.The2011Pacific-RimConferenceonMultimedia(PCM)2011. [9]LifangWu,LianchaoCao,JinqiaoWang,ShuqinLiu,ContentAwareMetricforImageResizingAssessment.The2011Pacific-RimConferenceonMultimedia(PCM)2011. [10]WeiFu,JinqiaoWang,XiaobinZhuHanqingLuandSongdema,VideoReshufflingwithNarrativestowardEffectiveVideoBrowsing,2011SixthInternationalConferenceonImageandGraphics(ICIG),pp821–826,2011. [11]XianbinZhu,JingLiu,JinqiaoWang,WeiFu,HanqingLu.GlobalTrajectoryConstructionacrossMulti-CamerasviaGraphMatching.2011SixthInternationalConferenceonImageandGraphics(ICIG),pp801–806,2011.

2010

[1]JianlongFu,JinqiaoWang,andHanqingLu,EffectiveLogoRetrievalwithAdaptiveLocalFeatureSelection.ACMMultimediaConference(MM),pp971-974,2010. [2]ShiChen,JinqiaoWang,YangLiu,ChangshengXu,HanqingLu.FastFeatureSelectionandTrainingforAdaBoost-basedConceptDetectionwithLargeScaleDatasets.ACMMultimediaConference(MM),pp1179-1182,2010. [3]XiaoXian,ChangshengXu,JinqiaoWang,LandmarkImageClassificationUsing3DPointClouds,ACMMultimediaConference(MM),pp821–826,2010. [4]ShiChen,JinqiaoWang,YiOuyang,BoWang,QiTian,HanqingLu.Multi-levelTrajectorymodelingForVideoCopyDetection.InternationalConferenceonAcoustics,Speech,andSignalProcessing(ICASSP),pp2378-2381,2010. [5]JieYang,JinqiaoWang,andHanqingLu.PeopleDetectionbyBoostingFeaturesinNonlinearSubspace.The2010Pacific-RimConferenceonMultimedia(PCM),pp227-235,2010. [6]ShiChen,JinqiaoWang,BoWang,Ling-YuDuan,QiTian,HanqingLu.AdVR:LinkingAdVideowithProductsorService.InternationalConferenceonMultimediaModeling(MMM),pp776-778,2010. [7]BoWang,JinqiaoWang,Ling-YuDuan,QiTian,HanqingLu,WenGao.InteractiveWebVideoAdvertisingwithContextAnalysisandSearch.IEEEconferenceonPatternRecognition(ICPR),pp3252-3255,2010. [8]BoWang,JinqiaoWang,Ling-YuDuan,QiTian,HanqingLu.InteractiveServiceRecommendationbasedonAdConceptHierarchy.InternationalConferenceonInternetMultimediaComputingandService(ICIMCS),pp87-90,2010. [9]LiliMa,JingLiu,JinqiaoWang,JianCheng,HanqingLu.AImprovedSilhouetteTrackingApproachIntegratingParticleFilterwithGraphCuts.InternationalConferenceonAcoustics,Speech,andSignalProcessing(ICASSP),pp1142-1145,2010. [10]LiliMa,JianCheng,JingLiu,JinqiaoWang,HanqingLu:VisualAttentionModelBasedObjectTracking.The2010Pacific-RimConferenceonMultimedia(PCM),pp483-493,2010. [11]ChaoLiang,YuJiang,JianCheng,ChangshengXu,XiaoweiLuo,JinqiaoWang,YuFu,HanqingLu,JianMa.PersonalizedSportsVideoCustomizationforMobileDevices.InternationalConferenceonMultimediaModeling(MMM),pp614-625,2010. [12]ChunjieZhang,JingLiu,JinqiaoWang,QiTian,ChangshengXu,HanqingLu,SongdeMa.ImageClassificationUsingSpatialPyramidCodingandVisualWordReweighting.ACCV,pp239-249,2010. [13]ChengliXie,JinqiaoWang,HanqingLu.KernelForestandItsApplicationonObjectDetection,The6thJointConferenceonHarmoniousHumanMachineEnvironment(HHME),pp1-5,2010. [14]JieYang,JinqiaoWang,andHanqingLu,AHierarchicalApproachforBackgroundModelingandMovingObjectsDetection.InternationalJournalofControl,Automation,andSystems,8(5),pp940-947,2010.