Dense Crowds Detection and Counting with a Lightweight Architecture

TL;DR

This paper introduces a lightweight CNN architecture for crowd detection and counting that balances resource efficiency with accuracy, suitable for embedded systems, and demonstrates its effectiveness on the USF-QNRF dataset.

Contribution

A novel lightweight CNN architecture trained with Bayes loss, pruned for efficiency, and validated on USF-QNRF, showing competitive accuracy with fewer parameters.

Findings

Achieved MAPE of 154.07 and MSE of 241.77 on USF-QNRF

Maintained only 0.067 million parameters, suitable for embedded devices

Bayes loss improves accuracy and the last convolutional layer may cause overfitting

Abstract

In the context of crowd counting, most of the works have focused on improving the accuracy without regard to the performance leading to algorithms that are not suitable for embedded applications. In this paper, we propose a lightweight convolutional neural network architecture to perform crowd detection and counting using fewer computer resources without a significant loss on count accuracy. The architecture was trained using the Bayes loss function to further improve its accuracy and then pruned to further reduce the computational resources used. The proposed architecture was tested over the USF-QNRF achieving a competitive Mean Average Error of 154.07 and a superior Mean Square Error of 241.77 while maintaining a competitive number of parameters of 0.067 Million. The obtained results suggest that the Bayes loss can be used with other architectures to further improve them and also the last convolutional layer provides no significant information and even encourage over-fitting at training.