ADCrowdNet: An Attention-injective Deformable Convolutional Network for Crowd Understanding

TL;DR

ADCrowdNet is a novel deep learning model that improves crowd counting accuracy in noisy, congested scenes by combining attention mechanisms with deformable convolutions, outperforming existing methods across multiple datasets.

Contribution

The paper introduces ADCrowdNet, a new attention-injective deformable convolutional network that enhances crowd understanding by effectively focusing on crowd regions and handling noise.

Findings

Outperforms state-of-the-art on four crowd datasets.

Effectively detects crowd regions and congestion levels.

More resistant to noise and congestion in crowd images.

Abstract

We propose an attention-injective deformable convolutional network called ADCrowdNet for crowd understanding that can address the accuracy degradation problem of highly congested noisy scenes. ADCrowdNet contains two concatenated networks. An attention-aware network called Attention Map Generator (AMG) first detects crowd regions in images and computes the congestion degree of these regions. Based on detected crowd regions and congestion priors, a multi-scale deformable network called Density Map Estimator (DME) then generates high-quality density maps. With the attention-aware training scheme and multi-scale deformable convolutional scheme, the proposed ADCrowdNet achieves the capability of being more effective to capture the crowd features and more resistant to various noises. We have evaluated our method on four popular crowd counting datasets (ShanghaiTech, UCF_CC_50, WorldEXPO'10, and UCSD) and an extra vehicle counting dataset TRANCOS, and our approach beats existing state-of-the-art approaches on all of these datasets.