中科院计算所视觉信息处理与学习组

您所在的位置 / 新闻动态

新闻动态

Global Perception GANs

发表日期:2017-05-26      点击击数: 137

Ran He
中国科学院自动化研究所


时间:5月26日上午10:00~12:00
地点:计算所446会议室


报告摘要:
Informationtheoryisthemathematicaltheoryofacquisition,representation,processing,andlearningofinformation.Inpatternrecognition,informationtheoryisthestudyofhowpatternrecognitionsystemsaccomplishthesetasks.Italsoexploitstheunderlyingpatternclassificationmechanicsandnewfeaturesofclassicalinformationtheory.Thisisanintroductorytalkoninformationtheoryforrobustpatternrecognition.ItwillshowourrecentworkstoimprovetherobustnessofGenerativeAdversarialNetworks(GANs).Particularly,globalandlocalconstraintsareconsideredinGANstoproducephotorealistic/visuallycompellingreconstructionresultsforanill-posedfacereconstructionproblem.


报告人简介:
RanHereceivedthePh.D.degreeinPatternRecognitionandIntelligentSystemsfromInstituteofAutomation,ChineseAcademyofSciences(CASIA)in2009.HeiscurrentlyaprofessorwithNationalLaboratoryofPatternRecognition(NLPR),CASIA.HeisalsoaProfessorwithUniversityofChineseAcademyofSciences.HehasobtainedseveralhonorsandawardssuchasGUCASLuJiaxiYoungTalentAwardandACMBeijingrisingstaraward.HeiscurrentlyaseniormemberofIEEE,servesasanassociateeditorofElsevierNeurocomputingandIETimageprocessing,andservesontheprogramcommitteeofseveralconferences.Hisresearchinterestsincludepatternrecognition,computervisionandinformationtheory.Hehaspublishedover100journalandconferencepapersinthesefields,andhaswidelypublishedathighlyrankedinternationaljournals,suchasIEEETPAMI,IEEETIP,IEEETNNLS,IEEETKDE,IEEETIFS,ElsevierPR,andMITNECO,andleadinginternationalconferences,suchasCVPR,ICCV,AAAI,IJCAIandSIGIR.

视觉信息处理和学习组
  • 单位地址:北京海淀区中关村科学院南路6号
  • 邮编:100190
  • 联系电话:010-62600514
  • Email:yi.cheng@vipl.ict.ac.cn
  • Valse

  • 深度学习大讲堂

版权所有 @ 中科院计算所视觉信息处理与学习组 京ICP备05002829号 京公网安备1101080060