Convolutional Neural Networks and Heuristic Methods for Crowd Counting: A Systematic Review

Abstract

The crowd counting task has become a pillar for crowd control as it provides information concerning the number of people in a scene. It is helpful in many scenarios such as video surveillance, public safety, and future event planning. To solve such tasks, researchers have proposed different solutions. In the beginning, researchers went with more traditional solutions, while recently the focus is on deep learning methods and, more specifically, on Convolutional Neural Networks (CNNs), because of their efficiency. This review explores these methods by focusing on their key differences, advantages, and disadvantages. We have systematically analyzed algorithms and works based on the different models suggested and the problems they are trying to solve. The main focus is on the shift made in the history of crowd counting methods, moving from the heuristic models to CNN models by identifying each category and discussing its different methods and architectures. After a deep study of the literature on crowd counting, the survey partitions current datasets into sparse and crowded ones. It discusses the reviewed methods by comparing their results on the different datasets. The findings suggest that the heuristic models could be even more effective than the CNN models in sparse scenarios.

Current approaches in UAV Operational Risk Assessment and Practical Considerations

“…Crowd counting, which estimates the number of people in images or videos, is a key aspect, and there are multiple techniques available for this task. There are five main crowd counting methods [5]: counting by detection, by regression, by density estimation, by clustering, and by CNN, the latter being the most accurate but computationally expensive. All of these methods rely on machine learning techniques.…”

Hierarchical Inverse Distance Transformer for Enhanced Localization in Dense Crowds

“…Contemporary crowd analysis methodologies primarily focus on counting, employing Convolutional Neural Networks (CNNs) [10,11] to create density maps that estimate crowd sizes through data synthesis. To increase the precision of these maps, various strategies merge features from disparate layers or scales.…”

Artificial Intelligence Aided Crowd Analytics in Rail Transit Station

“…To determine the crowd number in a high crowd density image with varied head sizes, CNN-based models with multi-column architecture have been proposed to extract multi-scale features (21), which improves the robustness of scale awareness (22). However, multicolumn architecture is bloated, and single-column architecture was proposed to make models simpler and more efficient (23,24). Furthermore, auxiliary-task models have been proposed, which can conduct one or more tasks related to crowd counting (22,24,25).…”