Crowd Transformer Network

TL;DR

This paper introduces a novel crowd counting method that combines local convolutional features with non-local self-attention features to improve crowd density estimation accuracy.

Contribution

It proposes integrating local and non-local features using convolution and self-attention mechanisms for enhanced crowd counting performance.

Findings

Significant improvement over previous methods on three datasets

Effective combination of local and non-local features

Demonstrates the usefulness of self-attention in crowd density estimation

Abstract

In this paper, we tackle the problem of Crowd Counting, and present a crowd density estimation based approach for obtaining the crowd count. Most of the existing crowd counting approaches rely on local features for estimating the crowd density map. In this work, we investigate the usefulness of combining local with non-local features for crowd counting. We use convolution layers for extracting local features, and a type of self-attention mechanism for extracting non-local features. We combine the local and the non-local features, and use it for estimating crowd density map. We conduct experiments on three publicly available Crowd Counting datasets, and achieve significant improvement over the previous approaches.