WhatMakesforEnd-to-EndObjectDetection?PeizeSun1YiJiang2EnzeXie1WenqiShao3ZehuanYuan2ChanghuWang2PingLuo1AbstractFigure1.End-to-endObjectdetection.Non-end-to-endObjectdetectorsrequireNMStoremoveredu...
RethinkingRotatedObjectDetectionwithGaussianWassersteinDistanceLossXueYang123JunchiYan12QiMing4WentaoWang1XiaopengZhang3QiTian3AbstractFigure1.ComparisonofthedetectionresultsbetweenSmoothL1loss-bas...
ObjectSegmentationWithoutLabelswithLarge-ScaleGenerativeModelsAndreyVoynov1StanislavMorozov1ArtemBabenko1Abstractmodelstoperformlabel-freeObjectsegmentation,wheregroundtruthpixel-levellabelsareexpe...
MoNet3D:TowardsAccurateMonocular3DObjectLocalizationinRealTimeXichuanZhou1YicongPeng1ChunqiaoLong1FengboRen2CongShi1AbstractRedmonetal.,2016;Liuetal.,2016).However,despiteitsgreatadvantagesintermso...
FrustratinglySimpleFew-ShotObjectDetectionXinWang1ThomasE.Huang2TrevorDarrell1JosephE.Gonzalez1FisherYu1Abstract2017;Gidaris&Komodakis,2018;Wangetal.,2019a)haveexploredtechniquestotransferknowledge...
Context-AwareZero-ShotLearningforObjectRecognitionE´loiZablocki1PatrickBordes1BenjaminPiwowarski1LaureSoulier1PatrickGallinari12Abstractaclassforwhichnosuperviseddataisavailable,byusingknowledgeac...
End-to-endActiveObjectTrackingviaReinforcementLearningWenhanLuo1PengSun1FangweiZhong2WeiLiu1TongZhang1YizhouWang2AbstractActionActionActiveTrackerCameraControlWestudyactiveObjecttracking,whereatrac...
DeepPredictiveCodingNetworkforObjectRecognitionHaiguangWen1KuanHan1JunxingShi1YizhenZhang1EugenioCulurciello12ZhongmingLiu12Abstractfeaturesbyafeedforwardpassthroughstacksofconvolu-tionallayers.Bas...