CACrowdGAN: Cascaded Attentional Generative Adversarial Network for Crowd Counting

Crowd counting is a valuable technology for extremely dense scenes in the transportation. Existing methods generally have higher-order inconsistencies between ground truth density maps and generated density maps. To address this issue, we incorporate an attentional discriminator to take charge of checking the density map between the generator and the ground truth. Thus, a Cascaded Attentional Generative Adversarial Network (CACrowdGAN) is proposed that enables the attentional-driven discriminator to distinguish implausible density maps and simultaneously to guide the generator to deliver fine-grained high quality density maps. The proposed CACrowdGAN consists of two components: an attentional generator and a cascaded attentional discriminator. The attentional generator has an attention module and a density module. The attention module is developed for the generator to focus on the crowd regions of the input images, while the density module is used to provide the attentional input of the discriminator. In addition, a cascaded attentional discriminator is proposed to synthesize attentional-driven fine-grained details at different crowd regions of the input image and compute a per-pixel fine-grained loss for training generator. The proposed CACrowdGAN achieves the state-of-the-art performance on five popular crowd counting datasets (ShanghaiTech, WorldEXPO’10, UCSD, UCF_CC_50 and UCF_QNRF), which demonstrates the effectiveness and robustness of the proposed approach in the complex scenes.

I. Introduction

With the rapid increase of the population of major cities, crowd analysis has drawn remarkable attention and already become an important security technique in video surveillance and intelligent transportation systems [1]–[8]. Crowd counting can help to improve the emergency planning and prevent congestion in train stations and airports. Various methods have been proposed to tackle this task [9]–[12]. Early works employ low-level features as region descriptors, followed by a classifier for classification [13], [14]. Benefiting from the recent progress in deep learning [15], [16], crowd counting approaches have seen a great success. However, due to problems such as the heavy occlusions and cluttered background, it still remains a challenging task in practical applications [17].