Scale-Context Perceptive Network for Crowd Counting and Localization in Smart City System

The task of crowd counting and localization is to predict the count and position of people in a crowd, which is a practical and essential subtask in crowd analysis and smart city systems. However, the inherent problems of scale variation and background disturbance restrain their performance. While recent research focus on studying counting and localization independently, a few works are capable of executing both tasks simultaneously. To this end, we propose a scale-context perceptive network (SCPNet) to jointly tackle the crowd counting and localization tasks in a unified framework. Specifically, a scale perceptive (SP) module with a local–global branch schema is designed to capture multiscale information. Meanwhile, a context perceptive (CP) module, by the channel-spatial self-attention mechanism, is derived to suppress the background disturbance. Furthermore, a novel hierarchical scale loss function that combines the Euclidean loss function and structural similarity loss function is designed to prompt the proposed model to fulfill the counting and localization simultaneously. Extensive experiments on challenging crowd data sets prove the superiority of the proposed SCPNet compared with the state-of-the-art competitors in both objective and subjective evaluations.

I. Introduction

Crowd analysis is an emerging topic in computer vision, and a crucial task in smart city applications, e.g., video monitoring, urban planning, and public security [1], [2]. It has two essential subtasks, namely, counting and localization, that have drawn signification attention in recent years. The objectives are to infer pedestrian numbers and locations, respectively. The approach for crowd counting is constantly refreshed and increasingly more effective. Meanwhile, crowd localization, evolved from crowd counting, is gradually explored and developed. They can be served for high-level vision tasks, e.g., crowd tracking [3] and 3-D human pose estimation [4].