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: Jul 16, 2026

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
11:54

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Published on: March 13, 2017

A Two-Stage Coarse-to-Fine Framework for Sparse Crowd Density Prediction in Digital Twin-Based Safety Monitoring.

Younghwan Jeong1, SoHyeon Kim1, Jinyoung Lee1

  • 1Korea Electronics Technology Institute, Seongnam-si 13509, Gyeonggi-do, Republic of Korea.

Sensors (Basel, Switzerland)
|July 15, 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

HCFL: hybrid contribution-driven federated learning for fair and efficient optimization.

Scientific reports·2026
Same author

Enhancing Ammonia Concentration Prediction with a Transfer-Learning-Based Model: Application in a Pig Farm.

Animals : an open access journal from MDPI·2026
Same author

Noise-Robust 3D Pose Estimation Using Appearance Similarity Based on the Distributed Multiple Views.

Sensors (Basel, Switzerland)·2024
Same author

Efficient Haze Removal from a Single Image Using a DCP-Based Lightweight U-Net Neural Network Model.

Sensors (Basel, Switzerland)·2024
Same author

An Implementation of Inverse Cosine Hardware for Sound Rendering Applications.

Sensors (Basel, Switzerland)·2023
Same author

A Cluster-Driven Adaptive Training Approach for Federated Learning.

Sensors (Basel, Switzerland)·2022

This study introduces a novel two-stage framework for proactive crowd congestion forecasting in digital twin platforms. It efficiently predicts hazardous crowd density by focusing computation on critical areas, improving accuracy and reducing resource use.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Public Safety

Background:

  • Crowd disasters in public spaces escalate rapidly, necessitating proactive rather than reactive safety measures.
  • Existing digital twin platforms require predictive modules to forecast crowd congestion before critical levels are reached.
  • Current AI models for crowd prediction are inefficient, applying uniform computation across all areas, which wastes resources and biases results.

Purpose of the Study:

  • To develop an efficient and accurate predictive module for digital twin platforms to forecast crowd congestion.
  • To address the limitations of conventional single-stage AI models in handling the spatio-temporal sparsity of crowd congestion.
  • To propose a coarse-to-fine framework that optimizes computational resources for crowd density prediction.
Keywords:
crowd density predictioncrowd safety monitoringdigital twinspatio-temporal sparsity

Related Experiment Videos

Last Updated: Jul 16, 2026

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles
11:54

Microfluidic Platform with Multiplexed Electronic Detection for Spatial Tracking of Particles

Published on: March 13, 2017

Main Methods:

  • A two-stage coarse-to-fine framework was developed, integrating CoarseSTFormer and SparseQueryDecoder modules.
  • The CoarseSTFormer identifies density-critical regions using low-resolution input, enabling efficient global screening.
  • The SparseQueryDecoder reconstructs high-resolution predictions selectively on identified candidate regions, avoiding uniform computation.

Main Results:

  • The proposed framework matches the strongest dense baselines in crowd congestion reconstruction quality.
  • It demonstrates a more balanced variance profile across different crowd density scales.
  • Significant reductions in GPU energy consumption (1.9×–5.0×) and computational cost (3.8×–54×) were achieved during inference.

Conclusions:

  • The developed framework offers a practical and resource-efficient solution for proactive crowd congestion forecasting in digital twins.
  • It effectively exploits the spatio-temporal sparsity of crowd congestion for improved prediction accuracy and efficiency.
  • This approach enhances the safety and operational management of dense public spaces by enabling timely interventions.