Cargando…

DMPNet: densely connected multi-scale pyramid networks for crowd counting

Crowd counting has been widely studied by deep learning in recent years. However, due to scale variation caused by perspective distortion, crowd counting is still a challenging task. In this paper, we propose a Densely Connected Multi-scale Pyramid Network (DMPNet) for count estimation and the gener...

Descripción completa

Detalles Bibliográficos
Autores principales: Li, Pengfei, Zhang, Min, Wan, Jian, Jiang, Ming
Formato: Online Artículo Texto
Lenguaje:English
Publicado: PeerJ Inc. 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9044264/
https://www.ncbi.nlm.nih.gov/pubmed/35494810
http://dx.doi.org/10.7717/peerj-cs.902
_version_ 1784695068165144576
author Li, Pengfei
Zhang, Min
Wan, Jian
Jiang, Ming
author_facet Li, Pengfei
Zhang, Min
Wan, Jian
Jiang, Ming
author_sort Li, Pengfei
collection PubMed
description Crowd counting has been widely studied by deep learning in recent years. However, due to scale variation caused by perspective distortion, crowd counting is still a challenging task. In this paper, we propose a Densely Connected Multi-scale Pyramid Network (DMPNet) for count estimation and the generation of high-quality density maps. The key component of our network is the Multi-scale Pyramid Network (MPN), which can extract multi-scale features of the crowd effectively while keeping the resolution of the input feature map and the number of channels unchanged. To increase the information transfer between the network layer, we used dense connections to connect multiple MPNs. In addition, we also designed a novel loss function, which can help our model achieve better convergence. To evaluate our method, we conducted extensive experiments on three challenging benchmark crowd counting datasets. Experimental results show that compared with the state-of-the-art algorithms, DMPNet performs well in both parameters and results. The code is available at: https://github.com/lpfworld/DMPNet.
format Online
Article
Text
id pubmed-9044264
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher PeerJ Inc.
record_format MEDLINE/PubMed
spelling pubmed-90442642022-04-28 DMPNet: densely connected multi-scale pyramid networks for crowd counting Li, Pengfei Zhang, Min Wan, Jian Jiang, Ming PeerJ Comput Sci Artificial Intelligence Crowd counting has been widely studied by deep learning in recent years. However, due to scale variation caused by perspective distortion, crowd counting is still a challenging task. In this paper, we propose a Densely Connected Multi-scale Pyramid Network (DMPNet) for count estimation and the generation of high-quality density maps. The key component of our network is the Multi-scale Pyramid Network (MPN), which can extract multi-scale features of the crowd effectively while keeping the resolution of the input feature map and the number of channels unchanged. To increase the information transfer between the network layer, we used dense connections to connect multiple MPNs. In addition, we also designed a novel loss function, which can help our model achieve better convergence. To evaluate our method, we conducted extensive experiments on three challenging benchmark crowd counting datasets. Experimental results show that compared with the state-of-the-art algorithms, DMPNet performs well in both parameters and results. The code is available at: https://github.com/lpfworld/DMPNet. PeerJ Inc. 2022-03-18 /pmc/articles/PMC9044264/ /pubmed/35494810 http://dx.doi.org/10.7717/peerj-cs.902 Text en ©2022 Li et al. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited.
spellingShingle Artificial Intelligence
Li, Pengfei
Zhang, Min
Wan, Jian
Jiang, Ming
DMPNet: densely connected multi-scale pyramid networks for crowd counting
title DMPNet: densely connected multi-scale pyramid networks for crowd counting
title_full DMPNet: densely connected multi-scale pyramid networks for crowd counting
title_fullStr DMPNet: densely connected multi-scale pyramid networks for crowd counting
title_full_unstemmed DMPNet: densely connected multi-scale pyramid networks for crowd counting
title_short DMPNet: densely connected multi-scale pyramid networks for crowd counting
title_sort dmpnet: densely connected multi-scale pyramid networks for crowd counting
topic Artificial Intelligence
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9044264/
https://www.ncbi.nlm.nih.gov/pubmed/35494810
http://dx.doi.org/10.7717/peerj-cs.902
work_keys_str_mv AT lipengfei dmpnetdenselyconnectedmultiscalepyramidnetworksforcrowdcounting
AT zhangmin dmpnetdenselyconnectedmultiscalepyramidnetworksforcrowdcounting
AT wanjian dmpnetdenselyconnectedmultiscalepyramidnetworksforcrowdcounting
AT jiangming dmpnetdenselyconnectedmultiscalepyramidnetworksforcrowdcounting