Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 27, 2026

Artificial Intelligence-Based System for Detecting Attention Levels in Students
06:37

Artificial Intelligence-Based System for Detecting Attention Levels in Students

Published on: December 15, 2023

Multi-scale pyramid fusion with overlap density attention module for crowd counting.

Avinash Rohra1, Baoqun Yin1, Aakash Kumar2

  • 1Department of Automation, University of Science and Technology of China, Hefei 230027, China.

Neural Networks : the Official Journal of the International Neural Network Society
|April 16, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Facilitators to and experience of psychological resilience during disease response among people with diabetes: a mixed-methods study using resilience framework.

Frontiers in psychology·2026
Same author

Urinary tract infection caused by rare yeast <i>Vanrija humicola</i> in a urothelial carcinoma patient: a case report and literature review.

Frontiers in immunology·2026
Same author

Correction: Urinary tract infection caused by rare yeast <i>Vanrija humicola</i> in a urothelial carcinoma patient: a case report and literature review.

Frontiers in immunology·2026
Same author

Targeting Cathepsin B with p-coumaric acid rescues WWP1-dependent proteasomal degradation of NLRP3 in hyperuricemic nephropathy.

Molecular medicine (Cambridge, Mass.)·2026
Same author

Combined Puerarin and Magnesium Acetyl Taurate Intervention Mitigates Autism-Like Pathology Through Glutamatergic and MAPK Pathway Regulation.

Neurochemical research·2026
Same author

Metamorphosis of the purple sea urchin Heliocidaris crassispina: Single-cell transcriptomic datasets.

Scientific data·2026

This study introduces a new Multi-Scale Pyramid Fusion with Overlap Density Attention (MSPF) network for accurate crowd counting. The MSPF model effectively handles dense scenes with occlusions and scale variations, improving public safety estimations.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Machine Learning

Background:

  • Crowd counting is vital for public safety but faces challenges like occlusion and scale variation in dense scenes.
  • Existing methods struggle with overlapping individuals and diverse crowd densities.

Purpose of the Study:

  • To propose a novel Multi-Scale Pyramid Fusion with Overlap Density Attention (MSPF) network.
  • To enhance crowd counting accuracy in challenging, high-density scenarios.

Main Methods:

  • Developed a framework with multi-scale pyramid fusion, overlap density attention, and feature enrichment modules.
  • Utilized an encoder for multi-scale feature extraction and an attention module to focus on informative regions.
  • Integrated multi-scale features adaptively and refined spatial information for density map generation.
Keywords:
Deep convolutionHigh-density crowdMulti-scale feature fusionOverlap crowd counting

Related Experiment Videos

Last Updated: Jun 27, 2026

Artificial Intelligence-Based System for Detecting Attention Levels in Students
06:37

Artificial Intelligence-Based System for Detecting Attention Levels in Students

Published on: December 15, 2023

Main Results:

  • The MSPF network demonstrated superior performance on the Highly-Packed-Crowd dataset and four benchmark datasets.
  • Achieved state-of-the-art results in accuracy, robustness, and efficiency for crowd estimation.
  • Effectively addressed challenges of severe occlusion and large-scale variations in dense crowds.

Conclusions:

  • The proposed MSPF network offers a significant advancement in crowd counting technology.
  • MSPF provides a robust and efficient solution for public safety applications in crowded environments.
  • The novel attention and fusion mechanisms are key to handling complex crowd scenes.