Revisiting Crowd Counting: State-of-the-art, Trends, and Future Perspectives

TL;DR

This paper provides a comprehensive, up-to-date review of deep learning methods for crowd counting, categorizing key contributions by architecture, learning, and evaluation, and ranking models by benchmark performance.

Contribution

It offers a systematic categorization and comparison of recent crowd counting methods, highlighting advancements and current state-of-the-art techniques.

Findings

Most recent models outperform earlier approaches on benchmark datasets.

Categorization by architecture, loss functions, and evaluation metrics enhances understanding.

The survey serves as a valuable resource for new researchers in the field.

Abstract

Crowd counting is an effective tool for situational awareness in public places. Automated crowd counting using images and videos is an interesting yet challenging problem that has gained significant attention in computer vision. Over the past few years, various deep learning methods have been developed to achieve state-of-the-art performance. The methods evolved over time vary in many aspects such as model architecture, input pipeline, learning paradigm, computational complexity, and accuracy gains etc. In this paper, we present a systematic and comprehensive review of the most significant contributions in the area of crowd counting. Although few surveys exist on the topic, our survey is most up-to date and different in several aspects. First, it provides a more meaningful categorization of the most significant contributions by model architectures, learning methods (i.e., loss functions), and evaluation methods (i.e., evaluation metrics). We chose prominent and distinct works and excluded similar works. We also sort the well-known crowd counting models by their performance over benchmark datasets. We believe that this survey can be a good resource for novice researchers to understand the progressive developments and contributions over time and the current state-of-the-art.