典型文献
Transferring priors from virtual data for crowd counting in real world
文献摘要：
In recent years,crowd counting has increasingly drawn attention due to its widespread applications in the field of computer vision.Most of the existing methods rely on datasets with scarce labeled images to train networks.They are prone to suffer from the over-fitting problem.Further,these existing datasets usually just give manually labeled annotations related to the head center position.This kind of annotation provides limited information.In this paper,we propose to exploit virtual synthetic crowd scenes to improve the perfor-mance of the counting network in the real world.Since we can obtain people masks easily in a synthetic dataset,we first learn to distinguish people from the background via a segmentation network using the synthetic data.Then we transfer the learned segmentation priors from synthetic data to real-world data.Finally,we train a density estimation network on real-world data by utilizing the obtained people masks.Our experiments on two crowd counting datasets demonstrate the effectiveness of the proposed method.
文献关键词：
中图分类号：
作者姓名：
Xiaoheng JIANG;Hao LIU;Li ZHANG;Geyang LI;Mingliang XU;Pei LV;Bing ZHOU
作者机构：
School of Information Engineering,Zhengzhou University,Zhengzhou 450001,China;School of Information Engineering,Henan University of Science and Technology,Luoyang 471000,China
文献出处：
引用格式：
[1]Xiaoheng JIANG;Hao LIU;Li ZHANG;Geyang LI;Mingliang XU;Pei LV;Bing ZHOU-.Transferring priors from virtual data for crowd counting in real world)[J].计算机科学前沿,2022(03):1-8
A类：
B类：
Transferring,priors,from,virtual,crowd,counting,real,world,In,recent,years,has,increasingly,drawn,attention,due,its,widespread,applications,field,computer,vision,Most,existing,methods,rely,datasets,scarce,labeled,images,train,networks,They,are,prone,suffer,over,fitting,problem,Further,these,usually,just,give,manually,annotations,related,head,center,position,This,kind,provides,limited,information,this,paper,we,exploit,synthetic,scenes,improve,perfor,mance,Since,can,people,masks,easily,first,distinguish,background,via,segmentation,using,Then,transfer,learned,Finally,density,estimation,by,utilizing,obtained,Our,experiments,demonstrate,effectiveness,proposed
AB值：
0.57895
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。
