Crowd counting at the edge using weighted knowledge distillation

<p dir="ltr">Visual crowd counting has gained serious attention during the last couple of years. The consistent contributions to this topic have now solved several inherited challenges such as scale variations, occlusions, and cross-scene applications. However, these works attempt to...

Full description

Saved in:
Bibliographic Details
Main Author: Muhammad Asif Khan (7367468) (author)
Other Authors: Hamid Menouar (16904844) (author), Ridha Hamila (7006457) (author), Adnan Abu-Dayya (16904850) (author)
Published: 2025
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1864513537552941056
author Muhammad Asif Khan (7367468)
author2 Hamid Menouar (16904844)
Ridha Hamila (7006457)
Adnan Abu-Dayya (16904850)
author2_role author
author
author
author_facet Muhammad Asif Khan (7367468)
Hamid Menouar (16904844)
Ridha Hamila (7006457)
Adnan Abu-Dayya (16904850)
author_role author
dc.creator.none.fl_str_mv Muhammad Asif Khan (7367468)
Hamid Menouar (16904844)
Ridha Hamila (7006457)
Adnan Abu-Dayya (16904850)
dc.date.none.fl_str_mv 2025-04-08T03:00:00Z
dc.identifier.none.fl_str_mv 10.1038/s41598-025-90750-5
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Crowd_counting_at_the_edge_using_weighted_knowledge_distillation/30306838
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Information and computing sciences
Computer vision and multimedia computation
Machine learning
Crowd counting
Lightweight models
Knowledge distillation
Deep learning
Computer vision
dc.title.none.fl_str_mv Crowd counting at the edge using weighted knowledge distillation
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Visual crowd counting has gained serious attention during the last couple of years. The consistent contributions to this topic have now solved several inherited challenges such as scale variations, occlusions, and cross-scene applications. However, these works attempt to improve accuracy and often ignore model size and computational complexity. Several practical applications employ resource-limited stand-alone devices like drones to run crowd models and require real-time inference. Though there have been some good efforts to develop lightweight shallow crowd models offering fast inference time, the relevant literature dedicated to lightweight crowd counting is limited. One possible reason is that lightweight deep-learning models suffer from accuracy degradation in complex scenes due to limited generalization capabilities. This paper addresses this important problem by proposing knowledge distillation to improve the learning capability of lightweight crowd models. Knowledge distillation enables lightweight models to emulate deeper models by distilling the knowledge learned by the deeper model during the training process. The paper presents a detailed experimental analysis with three lightweight crowd models over six benchmark datasets. The results report a clear significance of the proposed method supported by several ablation studies.</p><h2>Other Information</h2><p dir="ltr">Published in: Scientific Reports<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1038/s41598-025-90750-5" target="_blank">https://dx.doi.org/10.1038/s41598-025-90750-5</a></p>
eu_rights_str_mv openAccess
id Manara2_21da3df77a2a07d62687de1e96c809cb
identifier_str_mv 10.1038/s41598-025-90750-5
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30306838
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Crowd counting at the edge using weighted knowledge distillationMuhammad Asif Khan (7367468)Hamid Menouar (16904844)Ridha Hamila (7006457)Adnan Abu-Dayya (16904850)Information and computing sciencesComputer vision and multimedia computationMachine learningCrowd countingLightweight modelsKnowledge distillationDeep learningComputer vision<p dir="ltr">Visual crowd counting has gained serious attention during the last couple of years. The consistent contributions to this topic have now solved several inherited challenges such as scale variations, occlusions, and cross-scene applications. However, these works attempt to improve accuracy and often ignore model size and computational complexity. Several practical applications employ resource-limited stand-alone devices like drones to run crowd models and require real-time inference. Though there have been some good efforts to develop lightweight shallow crowd models offering fast inference time, the relevant literature dedicated to lightweight crowd counting is limited. One possible reason is that lightweight deep-learning models suffer from accuracy degradation in complex scenes due to limited generalization capabilities. This paper addresses this important problem by proposing knowledge distillation to improve the learning capability of lightweight crowd models. Knowledge distillation enables lightweight models to emulate deeper models by distilling the knowledge learned by the deeper model during the training process. The paper presents a detailed experimental analysis with three lightweight crowd models over six benchmark datasets. The results report a clear significance of the proposed method supported by several ablation studies.</p><h2>Other Information</h2><p dir="ltr">Published in: Scientific Reports<br>License: <a href="https://creativecommons.org/licenses/by/4.0" target="_blank">https://creativecommons.org/licenses/by/4.0</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1038/s41598-025-90750-5" target="_blank">https://dx.doi.org/10.1038/s41598-025-90750-5</a></p>2025-04-08T03:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1038/s41598-025-90750-5https://figshare.com/articles/journal_contribution/Crowd_counting_at_the_edge_using_weighted_knowledge_distillation/30306838CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/303068382025-04-08T03:00:00Z
spellingShingle Crowd counting at the edge using weighted knowledge distillation
Muhammad Asif Khan (7367468)
Information and computing sciences
Computer vision and multimedia computation
Machine learning
Crowd counting
Lightweight models
Knowledge distillation
Deep learning
Computer vision
status_str publishedVersion
title Crowd counting at the edge using weighted knowledge distillation
title_full Crowd counting at the edge using weighted knowledge distillation
title_fullStr Crowd counting at the edge using weighted knowledge distillation
title_full_unstemmed Crowd counting at the edge using weighted knowledge distillation
title_short Crowd counting at the edge using weighted knowledge distillation
title_sort Crowd counting at the edge using weighted knowledge distillation
topic Information and computing sciences
Computer vision and multimedia computation
Machine learning
Crowd counting
Lightweight models
Knowledge distillation
Deep learning
Computer vision