刘静 - 中国科学院自动化研究所 -

个人简介

招生专业 081104-模式识别与智能系统 081203-计算机应用技术招生方向网络图像分析、理解与检索基于深度学习的目标分类、检测与分割图像文本描述与视觉问答招生要求专业背景不限，但需对科研怀有强烈的兴趣，具有良好的数学基础、编程能力以及自主学习能力。由于招生数量有限，希望提前与我联系。另招收少量实习生，要求计算机相关专业的在读本科或研究生，具有较强的编程能力，实习期半年以上者优先。教育背景 2004-09--2008-01中国科学院自动化研究所博士 2001-09--2004-06山东大学硕士 1997-09--2001-06山东大学本科工作经历 2015-11--至今中国科学院自动化研究所模式识别国家重点实验室研究员 2010-11--2015-10中国科学院自动化研究所模式识别国家重点实验室副研究员 2008-01--2010-10中国科学院自动化研究所模式识别国家重点实验室助理研究员

研究领域

图像分析、理解与检索；移动媒体计算与应用；深度学习算法

近期论文

查看导师最新文章（温馨提示：请注意重名现象，建议点开原文通过作者单位确认）

（1）Improvingvisualquestionansweringusingdropoutandenhancedquestionencoder,PatternRecognition,2019,通讯作者（2）BTDP:TowardSparseFusionwithBlockTermDecompositionPoolingforVisualQuestionAnswering,ACMTransactionsonMultimediaComputing,CommunicationsandApplications,2019,通讯作者（3）DualAttentionNetworkforSceneSegmentation,CVPR,2019,通讯作者（4）Multi-StyleGAN:Multi-StyleImageCaptioningfromNon-ParallelDatausingGANsandBack-Translation,CVPR,2019,通讯作者（5）DenselyConnectedAttentionFlowforVisualQuestionAnswering,IJCAI,2019,通讯作者（6）EnhancingVisualQuestionAnsweringUsingDropout,ACMMultimedia,2018,第2作者（7）AnswerDistillationforVisualQuestionAnswering,AsianConferenceonComputerVision,2018,第2作者（8）CollaborativeDeconvolutionalNeuralNetworksforJointDepthEstimationandSemanticSegmentation,IEEEtransactionsonneuralnetworksandlearningsystems,2018,第1作者（9）HierarchicallySupervisedDeconvolutionalNetworkforSemanticVideoSegmentation,PatternRecognition,2017,通讯作者（10）Sketch-basedImageRetrievalusingGenerativeAdversarialNetworks,ACMMultimedia,2017,第2作者（11）Denselyconnecteddeconvolutionalnetworkforsemanticsegmentation,ICIP,2017,第2作者（12）Objectness-awareSemanticSegmentation,ACMMultimedia,2016,第2作者（13）Object-awareDeepNetworkforCommodityImageRetrieval,InternationalConferenceonMultimediaRetrieval,2016,第2作者（14）MultimediaNewsSummarizationinSearch,ACMTIST,2016,通讯作者（15）Chatwithillustration,MultimediaSystem,2016,通讯作者（16）Fine-grainedimageclassificationvialow-ranksparsecodingwithgeneralandclass-specificcodebooks,IEEEtransactionsonneuralnetworksandlearningsystems,2016,第3作者（17）Domain-SensitiveRecommendationwithUser-ItemSubgroupAnalysis,IEEETransactionsonKnowledgeDataEngineering,2016,第1作者（18）Objectco-segmentationviasalientandcommonregionsdiscovery,Neurocomputing,2016,通讯作者（19）PartiallySharedLatentFactorLearningwithMulti-ViewData,IEEETransactionsonNeuralNetworksandLearningSystems,2015,第1作者（20）Detectionguideddeconvolutionalnetworkforhierarchicalfeaturelearning,PatternRecognition,2015,第1作者（21）WeaklySupervisedRBMforSemanticSegmentation,InternationalJointConferenceonArtificialIntelligence,2015,第2作者（22）ColorNamesLearningUsingConvolutionalNeuralNetworks,InternationalConferenceonImageProcessing,2015,第2作者（23）Weakly-SupervisedImageSemanticSegmentationviaJointDualClusteringandDictionaryLearning,InternationalConferenceonImageProcessing,2015,第2作者（24）ExclusiveConstrainedDiscriminativeLearningforWeakly-SupervisedSemanticSegmentation,ACMMultimedia,2015,第2作者（25）Semi-andWeakly-SupervisedSemanticSegmentationwithDeepConvolutionalNeuralNetworks,ACMMultimedia,2015,第2作者（26）RobustStructuredSubspaceLearningforDataRepresentation,IEEETransactionsonPatternAnalysisandMachineIntelligence,2015,通讯作者（27）LearningRepresentativeandDiscriminativeImageRepresentationbyDeepAppearanceandSpatialCoding,ComputerVisionandImageUnderstanding,2015,通讯作者（28）ObjectCo-segmentationviaSalientandCommonRegionsDiscovery,Neurocomputing,2015,通讯作者（29）BoostedMIMLmethodforweakly-supervisedimagesemanticsegmentation,MultimediaToolsApplication,2015,通讯作者（30）PersonalizedGeo-SpecificTagRecommendationforPhotosonSocialWebsites,IEEETransactionsonMultimedia,2014,第1作者（31）LearningaRepresentativeandDiscriminativePartModelwithDeepConvolutionalFeaturesforSceneRecognition,ACCV,2014,第2作者（32）ImageRepresentationLearningbyDeepAppearanceandSpatialCoding,ACCV,2014,第2作者（33）LearningLow-RankRepresentationswithClasswiseBlock-DiagonalStructureforRobustFaceRecognition,AAAI,2014,第2作者（34）LabelingComplicatedObjects:Multi-ViewMulti-InstanceMulti-LabelLearning,AAAI,2014,第3作者（35）Sparsesemanticmetriclearningforimageretrieval,MultimediaSystem,2014,第1作者（36）Semi-supervisedUnifiedLatentFactorlearningwithmulti-viewdata,MachineVisionApplication,2014,通讯作者（37）LearningRobustFaceRepresentationWithClasswiseBlock-DiagonalStructure,IEEETransactionsonInformationForensicsandSecurity,2014,通讯作者（38）Clustering-GuidedSparseStructuralLearningforUnsupervisedFeatureSelection,IEEETransactionsonKnowledgeDataEngineering,2014,通讯作者（39）Adaptivespatialpartitionlearningforimageclassification,Neurocomputing,2014,通讯作者（40）ProjectiveMatrixFactorizationwithunifiedembeddingforsocialimagetagging,ComputerVisionandImageUnderstanding,2014,通讯作者（41）Imageclassificationbynon-negativesparsecoding,correlationconstrainedlow-rankandsparsedecomposition,ComputerVisionandImageUnderstanding,2014,通讯作者（42）Keyobservationselection-basedeffectivevideosynopsisforcameranetwork,MachineVisionApplication,2014,通讯作者（43）Sparserepresentationforrobustabnormalitydetectionincrowdedscenes,PatternRecognition,2014,通讯作者（44）ObjectCo-segmentationviaDiscriminativeLowrankMatrixRecovery,ACMMultimedia,2013,第2作者（45）Weakly-SupervisedDualClusteringforImageSemanticSegmentation,IEEEComputerVisionandPatternRecognition,2013,第2作者（46）CorrelationConsistencyConstrainedProbabilisticMatrixFactorizationforSocialTagRefinement,Neurocomputing,2012,第1作者（47）User-awareImageTagRefinementviaTernarySemanticAnalysis,IEEETransactionsonMultimedia,2012,通讯作者（48）BoostingSparsityConstrainedBi-LinearModelforObjectRecognition,IEEEMultimedia,2012,通讯作者（49）NoisyTagAlignmentwithImageRegions,IEEEInternationalConferenceonMultimediaandExpo.,2012,第2作者（50）UnsupervisedFeatureSelectionUsingNonnegativeSpectralAnalysis,AssociationfortheAdvancementofArtificialIntelligence,2012,第2作者（51）CorrelationMiningforNewsRetrieval,Chapter33forthebooktitledasSocialNetworks:ComputationalAspectsandMining,2012,第1作者（52）NonlinearMatrixFactorizationwithUnifiedEmbeddingforSocialTagRelevanceLearning,Neurocomputing,2012,通讯作者（53）OneStepBeyondBagsofFeatures:VisualCategorizationUsingComponents,IEEEInternationalConferenceonImageProcessing,2011,第1作者（54）BoostedExemplarLearningforActionRecognitionandAnnotation,IEEETransactionsonCircuitsandSystemsforVideoTechnology,2011,通讯作者（55）ImageClassificationbyNon-NegativeSparseCoding,Low-RankandSparseDecomposition,IEEEComputerVisionandPatternRecognition,2011,第2作者（56）User-awareImageTagRefinementviaTernaryAnalysis,ACMMultimediaConference,2011,第2作者（57）DescriptiveLocalFeatureGroupsforImageClassification,IEEEInternationalConferenceonImageProcessing,2011,第2作者（58）Cross-mediaretrieval:state-of-the-artandopenissues,InternationalJournalofMultimediaIntelligenceandSecurity,2010,第1作者（59）ImageAnnotationUsingMulti-CorrelationProbabilisticMatrixFactorization,ACMMultimediaConference,2010,第2作者（60）Multi-ModalMulti-CorrelationPerson-CentricNewsRetrieval,ACMConferenceonInformationandKnowledgeManagement,2010,第2作者（61）ImageClassificationUsingSpatialPyramidCodingandVisualWordReweighting,AsianConferenceonComputerVision,2010,第2作者（62）AGraph-BasedImageAnnotationFramework,PatternRecognitionLetters,2009,第1作者（63）ImageAnnotationviaMulti-GraphLearning,PatternRecognition,2009,第1作者（64）WebImageRetrievalviaLearningSemanticsofQueryImage,IEEEInternationalConferenceonMultimediaandExpo.,2009,第2作者（65）Categorysensitivecodebookconstructionforobjectcategoryrecognition,IEEEInternationalConferenceonImageProcessing,2009,第2作者（66）基于图学习的自动图像标注,计算机学报,2008,通讯作者