Performance Comparison and Analysis for Large-Scale Crowd Counting Based on Convolutional Neural Networks

Reem Alotaibi,Tarik Alafif,Ahmed Barnawi,Rui Wang,Bander Alzahrani,Long Hu

doi:10.1109/access.2020.3037395

Reem Alotaibi, Tarik Alafif + Show 4 more

Open Access

https://doi.org/10.1109/access.2020.3037395

Copy DOI

Abstract

Nowadays, crowd analysis is one of the most important concepts that needs be relied upon, it contributes to decision making and ensuring the safety and security of the crowd. There are a variety of interesting research problems within the scope of crowd analysis including crowd tracking, crowd behaviour recognition and crowd counting. Crowd counting based on images and videos has been studied in past years. Nonetheless, estimating and detecting the number of human heads remains a challenging task due to occlusions, resolution, and lighting changes. This paper provides an overview and performance comparison of crowd counting techniques using convolutional neural networks (CNN) based on density map estimation. In this paper, we present a comprehensive analysis and benchmarking of crowd counting based on the UCF-QNRF dataset that contains the largest number of crowd count images and head annotations available in the public domain. We also show the density maps generation and their empirical evaluation along with performance comparison.

Highlights

Crowd management of large-scale events has been under the spotlight of many research efforts
In this paper, we review and analyze crowd counting and convolutional neural networks (CNN)-based density estimation models
RELATED WORK This paper focuses on the estimation of the density map based on CNN and gets the number of people

Summary

Introduction

Crowd management of large-scale events has been under the spotlight of many research efforts. Such systems help for a proper pre-event planning in case of disasters and to ensure public safety and security. Crowd counting, and density estimation techniques are considered critical factors for any crowd management system. Various approaches have been proposed to address the crowd counting problem. These approaches are categorized into detection-based methods [1], [2], regression-based techniques [3], [4] and density estimation methods [5]–[7]

Methods

Results

Conclusion