典型文献
Domain Adaptive Semantic Segmentation via Entropy-Ranking and Uncertain Learning-Based Self-Training
文献摘要：
Dear Editor,
This letter develops two new self-training strategies for domain adaptive semantic segmentation,which formulate self-training into the processes of mining more training samples and reducing influence of the false pseudo-labels.Particularly,a self-training strategy based on entropy-ranking is proposed to mine intra-domain information.Thus,numerous false pseudo-labels can be exploited and rectified,and more pseudo-labels can be involved in training.Meanwhile,another novel self-training strategy is developed to handle the regions that may possess false pseudo-labels.In detail,a specific uncertain loss,that makes the network automatically decide whether the pseudo-labels are true,is proposed to improve the network optimization.Consequently,the influence of false pseudo-labels can be reduced.Experimental results prove that,compared with the baseline,the average mIoU performance gain brought by our method can attain 4.3％.Extensive benchmark experiments further highlight the effectiveness of our method against existing state-of-the-arts.
文献关键词：
中图分类号：
作者姓名：
Chengli Peng;Jiayi Ma
作者机构：
Electronic Information School,Wuhan University,Wuhan 430072,China
文献出处：
引用格式：
[1]Chengli Peng;Jiayi Ma-.Domain Adaptive Semantic Segmentation via Entropy-Ranking and Uncertain Learning-Based Self-Training)[J].自动化学报（英文版）,2022(08):1524-1527
A类：
B类：
Domain,Adaptive,Semantic,Segmentation,via,Entropy,Ranking,Uncertain,Learning,Based,Self,Training,Dear,Editor,
This,letter,develops,new,self,training,strategies,domain,adaptive,semantic,segmentation,which,formulate,into,processes,mining,more,samples,reducing,influence,false,pseudo,labels,Particularly,strategy,entropy,ranking,proposed,mine,intra,information,Thus,numerous,can,exploited,rectified,involved,Meanwhile,another,novel,developed,handle,regions,that,may,possess,In,detail,specific,uncertain,loss,makes,network,automatically,decide,whether,true,improve,optimization,Consequently,reduced,Experimental,results,compared,baseline,average,mIoU,performance,brought,by,our,method,attain,Extensive,benchmark,experiments,further,highlight,effectiveness,against,existing,state,arts
AB值：
0.648919
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
