Learn from Object Counting: Crowd Counting with Meta‐learning

Abstract The objective of crowd counting is to learn a counter that can estimate the number of people in a single image. So far, most of the proposed work evaluates the crowd density by fitting the constructed density map corresponding to the sample. The performance of those algorithms depends on a...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Changtong Zan, Baodi Liu, Weili Guan, Kai Zhang, Weifeng Liu
Formato:	article
Lenguaje:	EN
Publicado:	Wiley 2021
Materias:	Photography TR1-1050 Computer software QA76.75-76.765
Acceso en línea:	https://doaj.org/article/08fbf0dfe6774dd0b456101de27ed2da
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:08fbf0dfe6774dd0b456101de27ed2da
record_format	dspace
spelling	oai:doaj.org-article:08fbf0dfe6774dd0b456101de27ed2da2021-11-29T03:38:16ZLearn from Object Counting: Crowd Counting with Meta‐learning1751-96671751-965910.1049/ipr2.12241https://doaj.org/article/08fbf0dfe6774dd0b456101de27ed2da2021-12-01T00:00:00Zhttps://doi.org/10.1049/ipr2.12241https://doaj.org/toc/1751-9659https://doaj.org/toc/1751-9667Abstract The objective of crowd counting is to learn a counter that can estimate the number of people in a single image. So far, most of the proposed work evaluates the crowd density by fitting the constructed density map corresponding to the sample. The performance of those algorithms depends on a large amount of carefully prepared data. However, a significant problem with crowd data sets is the difficulty of labeling. To address such a situation, utilizing object counting data in few‐shot scenes is considered and an efficient algorithm to extract the meta‐information is proposed, thus improving the accuracy and convergence rate of the crowd counting tasks. Specifically, the counting network is trained with only object counting tasks constructed on different domains during the meta‐training phase. Then, the meta‐counter is testing on crowd counting tasks in the meta‐testing stage. Experimentally, it is demonstrated that the above way improves the converge rate and accuracy of crowd counting tasks on three crowd counting datasets when meta‐training on ten‐type object counting tasks.Changtong ZanBaodi LiuWeili GuanKai ZhangWeifeng LiuWileyarticlePhotographyTR1-1050Computer softwareQA76.75-76.765ENIET Image Processing, Vol 15, Iss 14, Pp 3543-3550 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Photography TR1-1050 Computer software QA76.75-76.765
spellingShingle	Photography TR1-1050 Computer software QA76.75-76.765 Changtong Zan Baodi Liu Weili Guan Kai Zhang Weifeng Liu Learn from Object Counting: Crowd Counting with Meta‐learning
description	Abstract The objective of crowd counting is to learn a counter that can estimate the number of people in a single image. So far, most of the proposed work evaluates the crowd density by fitting the constructed density map corresponding to the sample. The performance of those algorithms depends on a large amount of carefully prepared data. However, a significant problem with crowd data sets is the difficulty of labeling. To address such a situation, utilizing object counting data in few‐shot scenes is considered and an efficient algorithm to extract the meta‐information is proposed, thus improving the accuracy and convergence rate of the crowd counting tasks. Specifically, the counting network is trained with only object counting tasks constructed on different domains during the meta‐training phase. Then, the meta‐counter is testing on crowd counting tasks in the meta‐testing stage. Experimentally, it is demonstrated that the above way improves the converge rate and accuracy of crowd counting tasks on three crowd counting datasets when meta‐training on ten‐type object counting tasks.
format	article
author	Changtong Zan Baodi Liu Weili Guan Kai Zhang Weifeng Liu
author_facet	Changtong Zan Baodi Liu Weili Guan Kai Zhang Weifeng Liu
author_sort	Changtong Zan
title	Learn from Object Counting: Crowd Counting with Meta‐learning
title_short	Learn from Object Counting: Crowd Counting with Meta‐learning
title_full	Learn from Object Counting: Crowd Counting with Meta‐learning
title_fullStr	Learn from Object Counting: Crowd Counting with Meta‐learning
title_full_unstemmed	Learn from Object Counting: Crowd Counting with Meta‐learning
title_sort	learn from object counting: crowd counting with meta‐learning
publisher	Wiley
publishDate	2021
url	https://doaj.org/article/08fbf0dfe6774dd0b456101de27ed2da
work_keys_str_mv	AT changtongzan learnfromobjectcountingcrowdcountingwithmetalearning AT baodiliu learnfromobjectcountingcrowdcountingwithmetalearning AT weiliguan learnfromobjectcountingcrowdcountingwithmetalearning AT kaizhang learnfromobjectcountingcrowdcountingwithmetalearning AT weifengliu learnfromobjectcountingcrowdcountingwithmetalearning
_version_	1718407629121781760

Learn from Object Counting: Crowd Counting with Meta‐learning

Ejemplares similares