典型文献
Erroneous pixel prediction for semantic image segmentation
文献摘要：
We consider semantic image segmentation.Our method is inspired by Bayesian deep learning which improves image segmentation accuracy by modeling the uncertainty of the network output.In contrast to uncertainty,our method directly learns to predict the erroneous pixels of a segmentation network,which is modeled as a binary classification problem.It can speed up training comparing to the Monte Carlo integration often used in Bayesian deep learning.It also allows us to train a branch to correct the labels of erroneous pixels.Our method consists of three stages:(i)predict pixel-wise error probability of the initial result,(ii)redetermine new labels for pixels with high error probability,and(iii)fuse the initial result and the redetermined result with respect to the error probability.We formulate the error-pixel prediction problem as a classification task and employ an error-prediction branch in the network to predict pixel-wise error probabilities.We also introduce a detail branch to focus the training process on the erroneous pixels.We have experimentally validated our method on the Cityscapes and ADE20K datasets.Our model can be easily added to various advanced segmentation networks to improve their performance.Taking DeepLabv3+as an example,our network can achieve 82.88％of mIoU on Cityscapes testing dataset and 45.73％on ADE20K validation dataset,improving corresponding DeepLabv3+results by 0.74％and 0.13％respectively.
文献关键词：
中图分类号：
作者姓名：
Lixue Gong;Yiqun Zhang;Yunke Zhang;Yin Yang;Weiwei Xu
作者机构：
State Key Lab of CAD&CG,Zhejiang University,Hangzhou 310058,China;School of Computing Clemson University,South Carolina,29634,USA
文献出处：
引用格式：
[1]Lixue Gong;Yiqun Zhang;Yunke Zhang;Yin Yang;Weiwei Xu-.Erroneous pixel prediction for semantic image segmentation)[J].计算可视媒体（英文）,2022(01):165-175
A类：
Erroneous,redetermine,redetermined,DeepLabv3+as,DeepLabv3+results
B类：
prediction,semantic,image,segmentation,We,consider,Our,method,inspired,by,Bayesian,deep,learning,which,improves,accuracy,modeling,uncertainty,output,In,contrast,our,directly,learns,erroneous,pixels,modeled,binary,classification,problem,It,can,speed,up,training,comparing,Monte,Carlo,integration,often,used,also,allows,branch,correct,labels,consists,three,stages,wise,error,probability,initial,new,high,iii,fuse,formulate,task,employ,probabilities,introduce,detail,focus,process,have,experimentally,validated,Cityscapes,ADE20K,datasets,easily,added,various,advanced,networks,their,performance,Taking,example,achieve,mIoU,testing,validation,improving,corresponding,respectively
AB值：
0.439981
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。
