Pedestrian dynamics modeling and social force analysis based on object detection

IntroductionObject detection is a fundamental component of modern computational applications, playing a crucial role in pedestrian analysis, autonomous navigation, and crowd monitoring. Despite its widespread utility, pedestrian-oriented object detection faces significant challenges, including dynam...

Full description

Saved in:
Bibliographic Details
Published in:Frontiers in physics Vol. 13
Main Authors: Yang, Fangyan, Liu, Rong, Zhu, Daoyu
Format: Journal Article
Language:English
Published: Frontiers Media S.A 07.07.2025
Subjects:
ISSN:2296-424X, 2296-424X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract IntroductionObject detection is a fundamental component of modern computational applications, playing a crucial role in pedestrian analysis, autonomous navigation, and crowd monitoring. Despite its widespread utility, pedestrian-oriented object detection faces significant challenges, including dynamic crowd behaviors, occlusions, multi-scale variability, and complex urban environments, which hinder the accuracy and robustness of existing models.MethodsTo address these challenges, we propose a novel framework that integrates the Information-Geometric Variational Inference Framework (IGVIF) with the Adaptive Exploration-Exploitation Trade-off Strategy (AEETS), specifically tailored for pedestrian dynamics. IGVIF formulates pedestrian detection as a probabilistic inference problem, leveraging principles from information geometry to efficiently explore high-dimensional parameter spaces. By incorporating techniques such as Riemannian optimization and multi-scale parameterization, IGVIF effectively captures the hierarchical and multi-modal structures inherent in pedestrian movement patterns. AEETS dynamically balances global exploration with local refinement using entropy-based metrics and feedback-driven adjustments, allowing the system to adaptively optimize complex loss landscapes with greater precision in pedestrian scenarios.ResultsTogether, these components create a robust and adaptive framework that overcomes traditional limitations by efficiently handling large-scale pedestrian variability and densely populated environments. Experimental evaluations across multiple real-world pedestrian datasets demonstrate the superiority of our physics-inspired approach, achieving state-of-the-art performance in pedestrian detection and movement analysis.DiscussionThis work highlights the transformative potential of interdisciplinary strategies in advancing pedestrian-aware object detection, bridging computational physics with deep learning methodologies to enhance urban mobility and crowd safety.
AbstractList IntroductionObject detection is a fundamental component of modern computational applications, playing a crucial role in pedestrian analysis, autonomous navigation, and crowd monitoring. Despite its widespread utility, pedestrian-oriented object detection faces significant challenges, including dynamic crowd behaviors, occlusions, multi-scale variability, and complex urban environments, which hinder the accuracy and robustness of existing models.MethodsTo address these challenges, we propose a novel framework that integrates the Information-Geometric Variational Inference Framework (IGVIF) with the Adaptive Exploration-Exploitation Trade-off Strategy (AEETS), specifically tailored for pedestrian dynamics. IGVIF formulates pedestrian detection as a probabilistic inference problem, leveraging principles from information geometry to efficiently explore high-dimensional parameter spaces. By incorporating techniques such as Riemannian optimization and multi-scale parameterization, IGVIF effectively captures the hierarchical and multi-modal structures inherent in pedestrian movement patterns. AEETS dynamically balances global exploration with local refinement using entropy-based metrics and feedback-driven adjustments, allowing the system to adaptively optimize complex loss landscapes with greater precision in pedestrian scenarios.ResultsTogether, these components create a robust and adaptive framework that overcomes traditional limitations by efficiently handling large-scale pedestrian variability and densely populated environments. Experimental evaluations across multiple real-world pedestrian datasets demonstrate the superiority of our physics-inspired approach, achieving state-of-the-art performance in pedestrian detection and movement analysis.DiscussionThis work highlights the transformative potential of interdisciplinary strategies in advancing pedestrian-aware object detection, bridging computational physics with deep learning methodologies to enhance urban mobility and crowd safety.
Author Yang, Fangyan
Zhu, Daoyu
Liu, Rong
Author_xml – sequence: 1
  givenname: Fangyan
  surname: Yang
  fullname: Yang, Fangyan
– sequence: 2
  givenname: Rong
  surname: Liu
  fullname: Liu, Rong
– sequence: 3
  givenname: Daoyu
  surname: Zhu
  fullname: Zhu, Daoyu
BookMark eNpNkMtKAzEYhYNUsNY-gLu8QGtuk0mWUrwUCnah4C7k8qemTCdlMpt5e2e0iJtzDmfxLb5bNGtzCwjdU7LmXOmHeP4a1oywak2rWjNFrtCcMS1XgonP2b99g5alHAkhlFVaMTFH-z0EKH2XbIvD0NpT8gWfcoAmtQds24BL9sk2OObOw3jYZiipYGcLBJxbnN0RfI8D9GOl3N6h62ibAstLL9DH89P75nW1e3vZbh53K8-J6qfUUTsCCiqnBFAVQUZe147aqLVWwKUDEW0AwYDVFUhOgXIlQdXj5gu0_eWGbI_m3KWT7QaTbTI_R-4OxnZ98g0YaZUjtfMepBZSShsZE1WlCNFc10GNLPrL8l0upYP4x6PETIbNZNhMhs3FMP8GkmNxug
Cites_doi 10.1016/j.patcog.2020.107404
10.1186/s12879-023-08795-8
10.1109/tgrs.2023.3348555
10.1109/ICCV48922.2021.00349
10.1371/journal.pone.0024391
10.1109/ICCV48922.2021.00350
10.1016/j.asoc.2024.112536
10.1109/ICCV48922.2021.00305
10.1103/physreve.102.052314
10.1109/tpami.2021.3085766
10.3389/fphy.2025.1522172
10.1109/CVPR46437.2021.00577
10.1109/cvpr52688.2022.00571
10.1016/j.jag.2024.104208
10.1016/j.bspc.2023.105941
10.3389/fphy.2024.1508465
10.1016/j.jvoice.2024.09.002
10.1109/CVPR46437.2021.00281
10.3389/fphy.2024.1529899
10.1007/978-3-031-20077-9_17
10.1109/ICCVW54120.2021.00107
10.1016/j.ipm.2023.103584
10.1007/978-3-030-58452-8_13
10.18127/j00338486-202109-11
10.1109/ICCVW54120.2021.00312
10.1016/j.physa.2009.03.023
10.1109/ICCV48922.2021.00294
10.1007/978-3-031-19812-0_31
10.3390/electronics12102323
10.3390/s23167190
10.1016/j.inpa.2023.02.007
10.1109/ICCV48922.2021.00290
ContentType Journal Article
DBID AAYXX
CITATION
DOA
DOI 10.3389/fphy.2025.1579280
DatabaseName CrossRef
DOAJ Directory of Open Access Journals
DatabaseTitle CrossRef
DatabaseTitleList
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
DeliveryMethod fulltext_linktorsrc
Discipline Physics
EISSN 2296-424X
ExternalDocumentID oai_doaj_org_article_6a8b07bcce694666af224558009397d8
10_3389_fphy_2025_1579280
GroupedDBID 5VS
9T4
AAFWJ
AAYXX
ACGFS
ADBBV
ADMLS
AFPKN
ALMA_UNASSIGNED_HOLDINGS
ARCSS
BCNDV
CITATION
GROUPED_DOAJ
KQ8
M~E
OK1
ID FETCH-LOGICAL-c308t-c309f9b0e8e5b84e18fe6f377b1af9998e36be4fade42e275e631e1386e87e633
IEDL.DBID DOA
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001531589400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 2296-424X
IngestDate Fri Oct 03 12:40:31 EDT 2025
Sat Nov 29 07:48:35 EST 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c308t-c309f9b0e8e5b84e18fe6f377b1af9998e36be4fade42e275e631e1386e87e633
OpenAccessLink https://doaj.org/article/6a8b07bcce694666af224558009397d8
ParticipantIDs doaj_primary_oai_doaj_org_article_6a8b07bcce694666af224558009397d8
crossref_primary_10_3389_fphy_2025_1579280
PublicationCentury 2000
PublicationDate 2025-07-07
PublicationDateYYYYMMDD 2025-07-07
PublicationDate_xml – month: 07
  year: 2025
  text: 2025-07-07
  day: 07
PublicationDecade 2020
PublicationTitle Frontiers in physics
PublicationYear 2025
Publisher Frontiers Media S.A
Publisher_xml – name: Frontiers Media S.A
References Gu (B19) 2021
Lou (B14) 2023; 12
Zhu (B2) 2020
Dmitriev (B32) 2025; 12
Zhang (B44) 2024
Sheng (B28) 2025; 13
Chen (B40) 2024; 134
Luo (B36) 2025; 12
Zhang (B41) 2025
Guan (B46) 2024; 61
Lv (B4) 2023
Wang (B22) 2021
Feng (B33) 2021
Li (B8) 2022
Misra (B26) 2021
Ishwerlal (B37) 2024; 91
Wang (B15) 2023
Xu (B21) 2021
Liu (B34) 2021
Sun (B23) 2021
Han (B29) 2021
Fu (B42) 2024; 11
Wang (B16) 2021
Liu (B17) 2021
Liu (B3) 2023
Liu (B12) 2022
Xie (B20) 2021
Bai (B11) 2022
Qin (B18) 2020; 106
Mao (B38) 2024; 169
Ng (B43) 2024; 24
Joseph (B24) 2021
Zhou (B31) 2011; 6
Zhang (B7) 2022
Fan (B25) 2021; 44
Yin (B6) 2020
Chen (B39) 2024; 62
Virasova (B5) 2021
Carion (B1) 2020
Zhu (B9) 2021
Zhou (B27) 2020; 102
Liu (B13) 2022
Reading (B30) 2021
Li (B10) 2022
Cai (B45) 2024
Zhou (B35) 2009; 388
References_xml – volume: 106
  start-page: 107404
  year: 2020
  ident: B18
  article-title: U2-net: going deeper with nested u-structure for salient object detection
  publication-title: Pattern Recognition
  doi: 10.1016/j.patcog.2020.107404
– volume: 24
  start-page: 123
  year: 2024
  ident: B43
  article-title: Study protocol: infectious diseases consortium (i3d) for study on integrated and innovative approaches for management of respiratory infections: respiratory infections research and outcome study (respiro)
  publication-title: BMC Infect Dis
  doi: 10.1186/s12879-023-08795-8
– volume: 62
  start-page: 1
  year: 2024
  ident: B39
  article-title: Dtssnet: dynamic training sample selection network for uav object detection
  publication-title: IEEE Trans Geosci Remote Sensing
  doi: 10.1109/tgrs.2023.3348555
– volume-title: IEEE international conference on computer vision
  year: 2021
  ident: B33
  article-title: Tood: task-aligned one-stage object detection
  doi: 10.1109/ICCV48922.2021.00349
– volume: 6
  start-page: e24391
  year: 2011
  ident: B31
  article-title: Investment strategies used as spectroscopy of financial markets reveal new stylized facts
  publication-title: PloS one
  doi: 10.1371/journal.pone.0024391
– volume-title: Detrs beat yolos on real-time object detection
  year: 2023
  ident: B4
– start-page: 1
  volume-title: 2024 IEEE international joint Conference on biometrics (IJCB)
  year: 2024
  ident: B44
  article-title: Blip-adapter: bridging vision-language models with adapters for generalizable face anti-spoofing
– volume-title: IEEE international conference on computer vision
  year: 2021
  ident: B20
  article-title: Oriented r-cnn for object detection
  doi: 10.1109/ICCV48922.2021.00350
– volume: 169
  start-page: 112536
  year: 2024
  ident: B38
  article-title: Medical supervised masked autoencoder: crafting a better masking strategy and efficient fine-tuning schedule for medical image classification
  publication-title: Appl Soft Comput
  doi: 10.1016/j.asoc.2024.112536
– volume-title: Transfusion: robust lidar-camera fusion for 3d object detection with transformers
  year: 2022
  ident: B11
– volume-title: IEEE international conference on computer vision
  year: 2021
  ident: B21
  article-title: End-to-end semi-supervised object detection with soft teacher
  doi: 10.1109/ICCV48922.2021.00305
– volume-title: European conference on computer vision
  year: 2023
  ident: B3
  article-title: Grounding dino: marrying dino with grounded pre-training for open-set object detection
– volume: 102
  start-page: 052314
  year: 2020
  ident: B27
  article-title: Predicting highway freight transportation networks using radiation models
  publication-title: Phys Rev E
  doi: 10.1103/physreve.102.052314
– volume: 44
  start-page: 6024
  year: 2021
  ident: B25
  article-title: Concealed object detection
  publication-title: IEEE Trans Pattern Anal Machine Intelligence
  doi: 10.1109/tpami.2021.3085766
– volume: 13
  start-page: 1522172
  year: 2025
  ident: B28
  article-title: Research on the integration of mems and reliable transmission of deep space networks based on time-sensitive networking
  publication-title: Front Phys
  doi: 10.3389/fphy.2025.1522172
– volume-title: Towards open world object detection
  year: 2021
  ident: B24
  doi: 10.1109/CVPR46437.2021.00577
– volume-title: International conference on learning representations
  year: 2020
  ident: B2
  article-title: Deformable detr: deformable transformers for end-to-end object detection
– volume-title: International conference on learning representations
  year: 2022
  ident: B7
  article-title: Dino: detr with improved denoising anchor boxes for end-to-end object detection
– start-page: 5792
  year: 2022
  ident: B13
  article-title: Target-aware dual adversarial learning and a multi-scenario multi-modality benchmark to fuse infrared and visible for object detection
  publication-title: Computer Vis Pattern Recognition
  doi: 10.1109/cvpr52688.2022.00571
– volume: 134
  start-page: 104208
  year: 2024
  ident: B40
  article-title: Object detection in aerial images using dota dataset: a survey
  publication-title: Int J Appl Earth Observation Geoinformation
  doi: 10.1016/j.jag.2024.104208
– volume-title: AAAI conference on artificial intelligence
  year: 2022
  ident: B8
  article-title: Bevdepth: acquisition of reliable depth for multi-view 3d object detection
– volume: 91
  start-page: 105941
  year: 2024
  ident: B37
  article-title: Lung disease classification using chest x ray image: an optimal ensemble of classification with hybrid training
  publication-title: Biomed Signal Process Control
  doi: 10.1016/j.bspc.2023.105941
– volume-title: Conference on robot learning
  year: 2021
  ident: B16
  article-title: Detr3d: 3d object detection from multi-view images via 3d-to-2d queries
– volume: 12
  start-page: 1508465
  year: 2025
  ident: B32
  article-title: Self-organization of the stock exchange to the edge of a phase transition: empirical and theoretical studies
  publication-title: Front Phys
  doi: 10.3389/fphy.2024.1508465
– year: 2024
  ident: B45
  article-title: Voice disorder classification using wav2vec 2.0 feature extraction
  publication-title: J Voice
  doi: 10.1016/j.jvoice.2024.09.002
– volume-title: Center-based 3d object detection and tracking
  year: 2020
  ident: B6
– volume-title: Computer vision and pattern recognition
  year: 2021
  ident: B29
  article-title: Redet: a rotation-equivariant detector for aerial object detection
  doi: 10.1109/CVPR46437.2021.00281
– volume: 12
  start-page: 1529899
  year: 2025
  ident: B36
  article-title: Application of mems technology in anti-electromagnetic radiation maternity clothes: state of the art and future perspectives
  publication-title: Front Phys
  doi: 10.3389/fphy.2024.1529899
– year: 2021
  ident: B30
  article-title: Categorical depth distribution network for monocular 3d object detection
  publication-title: Computer Vis Pattern Recognition
– volume-title: European conference on computer vision
  year: 2022
  ident: B10
  article-title: Exploring plain vision transformer backbones for object detection
  doi: 10.1007/978-3-031-20077-9_17
– volume-title: Fsce: few-shot object detection via contrastive proposal encoding
  year: 2021
  ident: B23
– start-page: 310
  volume-title: European conference on computer vision
  year: 2025
  ident: B41
  article-title: Long-clip: unlocking the long-text capability of clip
– volume-title: 2021 IEEE/CVF international conference on computer vision workshops (ICCVW)
  year: 2021
  ident: B22
  article-title: Fcos3d: fully convolutional one-stage monocular 3d object detection
  doi: 10.1109/ICCVW54120.2021.00107
– volume: 61
  start-page: 103584
  year: 2024
  ident: B46
  article-title: A t5-based interpretable reading comprehension model with more accurate evidence training
  publication-title: Inf Process & Management
  doi: 10.1016/j.ipm.2023.103584
– volume-title: European conference on computer vision
  year: 2020
  ident: B1
  article-title: End-to-end object detection with transformers
  doi: 10.1007/978-3-030-58452-8_13
– start-page: 115
  year: 2021
  ident: B5
  article-title: Rich feature hierarchies for accurate object detection and semantic segmentation
  publication-title: Radioengineering
  doi: 10.18127/j00338486-202109-11
– volume-title: 2021 IEEE/CVF international conference on computer vision workshops (ICCVW)
  year: 2021
  ident: B9
  article-title: Tph-yolov5: improved yolov5 based on transformer prediction head for object detection on drone-captured scenarios
  doi: 10.1109/ICCVW54120.2021.00312
– volume: 388
  start-page: 2623
  year: 2009
  ident: B35
  article-title: Numerical investigations of discrete scale invariance in fractals and multifractal measures
  publication-title: Physica A: Stat Mech its Appl
  doi: 10.1016/j.physa.2009.03.023
– volume-title: IEEE international conference on computer vision
  year: 2021
  ident: B34
  article-title: Group-free 3d object detection via transformers
  doi: 10.1109/ICCV48922.2021.00294
– volume-title: European conference on computer vision
  year: 2022
  ident: B12
  article-title: Petr: position embedding transformation for multi-view 3d object detection
  doi: 10.1007/978-3-031-19812-0_31
– volume: 12
  start-page: 2323
  year: 2023
  ident: B14
  article-title: Dc-yolov8: small-size object detection algorithm based on camera sensor
  publication-title: Electronics
  doi: 10.3390/electronics12102323
– volume-title: International conference on learning representations
  year: 2021
  ident: B17
  article-title: Unbiased teacher for semi-supervised object detection
– volume-title: Italian national conference on sensors
  year: 2023
  ident: B15
  article-title: Uav-yolov8: a small-object-detection model based on improved yolov8 for uav aerial photography scenarios
  doi: 10.3390/s23167190
– volume-title: International conference on learning representations
  year: 2021
  ident: B19
  article-title: Open-vocabulary object detection via vision and language knowledge distillation
– volume: 11
  start-page: 249
  year: 2024
  ident: B42
  article-title: Crop pest image recognition based on the improved vit method
  publication-title: Inf Process Agric
  doi: 10.1016/j.inpa.2023.02.007
– volume-title: IEEE international conference on computer vision
  year: 2021
  ident: B26
  article-title: An end-to-end transformer model for 3d object detection
  doi: 10.1109/ICCV48922.2021.00290
SSID ssj0001259824
Score 2.303812
Snippet IntroductionObject detection is a fundamental component of modern computational applications, playing a crucial role in pedestrian analysis, autonomous...
SourceID doaj
crossref
SourceType Open Website
Index Database
SubjectTerms crowd dynamics
deep learning
pedestrian detection
social force model
variational inference
Title Pedestrian dynamics modeling and social force analysis based on object detection
URI https://doaj.org/article/6a8b07bcce694666af224558009397d8
Volume 13
WOSCitedRecordID wos001531589400001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVAON
  databaseName: DOAJ Directory of Open Access Journals
  customDbUrl:
  eissn: 2296-424X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001259824
  issn: 2296-424X
  databaseCode: DOA
  dateStart: 20130101
  isFulltext: true
  titleUrlDefault: https://www.doaj.org/
  providerName: Directory of Open Access Journals
– providerCode: PRVHPJ
  databaseName: ROAD: Directory of Open Access Scholarly Resources
  customDbUrl:
  eissn: 2296-424X
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001259824
  issn: 2296-424X
  databaseCode: M~E
  dateStart: 20130101
  isFulltext: true
  titleUrlDefault: https://road.issn.org
  providerName: ISSN International Centre
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV07S8RAEF5EFGzEJ54vtrAS4u1mN_soVe6w8bhC4bqwj1mwycldtPS3u48o19nYDGEJYfgmMPNNJt8gdBOErK0mpKIh6IobZyrTCFVxbbwWzkRL8rIJOZupxULPN1Z9pZmwIg9cgBsLoyyR1jkQSQpdmBCTTtOoRMW19Pk3XyL1Bpkq3ZUkTMfLZ8zIwvQ4RK8jHaybO9pIXScZyI1EtKHXnxPL9ADtDxUhvi-eHKIt6I7Qbp7MdOtjNJ-Dh7xco8O-rI9f47y_JiYdbDqPS9sbx-rTQTwoIiM4pSePlx1e2tRqwR76PHXVnaDX6eTl8aka1iBUjhHVJ6uDtgQUNFZxoCqACExKS02I9Z0CJizwYDzwGmrZgGAUKFMClIzX7BRtd8sOzhBWjBpDnPDRcA7MRnIcEzb3opZaOjpCtz-YtO9F7aKNLCEB2CYA2wRgOwA4Qg8Jtd8bk1B1Pojha4fwtX-F7_w_HnKB9pJjeYpWXqLtfvUBV2jHffZv69V1fjOiff6afANVK74R
linkProvider Directory of Open Access Journals
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Pedestrian+dynamics+modeling+and+social+force+analysis+based+on+object+detection&rft.jtitle=Frontiers+in+physics&rft.au=Yang%2C+Fangyan&rft.au=Liu%2C+Rong&rft.au=Zhu%2C+Daoyu&rft.date=2025-07-07&rft.issn=2296-424X&rft.eissn=2296-424X&rft.volume=13&rft_id=info:doi/10.3389%2Ffphy.2025.1579280&rft.externalDBID=n%2Fa&rft.externalDocID=10_3389_fphy_2025_1579280
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2296-424X&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2296-424X&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2296-424X&client=summon