STNet: Scale Tree Network With Multi-Level Auxiliator for Crowd Counting

State-of-the-art approaches for crowd counting resort to deepneural networks to predict density maps. However, counting people in congested scenes remains a challenging task because the presence of drastic scale variation, density inconsistency, and complex background can seriously degrade their counting accuracy. To battle the ingrained issue of accuracy degradation, in this paper, we propose a novel and powerful network called Scale Tree Network (STNet) for accurate crowd counting. STNet consists of two key components: a Scale-Tree Diversity Enhancer and a Multi-level Auxiliator. Specifically, the Diversity Enhancer is designed to enrich scale diversity, which alleviates limitations of existing methods caused by insufficient level of scales. A novel tree structure is adopted to hierarchically parse coarse-to-fine crowd regions. Furthermore, a simple yet effective Multi-level Auxiliator is presented to aid in exploiting generalisable shared characteristics at multiple levels, allowing more accurate pixel-wise background cognition. The overall STNet is trained in an end-to-end manner, without the needs for manually tuning loss weights between the main and the auxiliary tasks. Extensive experiments on five challenging crowd datasets demonstrate the superiority of the proposed method.

I. Introduction

Crowd counting recently has drawn lots of attention from researchers due to its great importance in a wide array of real-world applications including video surveillance, public crowd monitoring, and traffic control. The main objective of crowd counting is to infer the number of people in congested images. Despite the exploration of pioneer works [1]–[6], crowd understanding is still a challenging issue for scenes exhibiting drastic scale variations, density inconsistency, or complex background.