Rethinking Global Context in Crowd Counting

Abstract

This paper investigates the role of global context for crowd counting. Specifically, a pure transformer is used to extract features with global information from overlapping image patches. Inspired by classification, we add a context token to the input sequence, to facilitate information exchange with tokens corresponding to image patches throughout transformer layers. Due to the fact that transformers do not explicitly model the tried-and-true channel-wise interactions, we propose a token-attention module (TAM) to recalibrate encoded features through channel-wise attention informed by the context token. Beyond that, it is adopted to predict the total person count of the image through regression-token module (RTM). Extensive experiments on various datasets, including ShanghaiTech, UCF-QNRF, JHU-CROWD++ and NWPU, demonstrate that the proposed context extraction techniques can significantly improve the performance over the baselines.

Accepted by Machine Intelligence Research (MIR)

References in corpus (5)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• A Survey of Recent Advances in CNN-based Single Image Crowd Counting and Density Estimation
• NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localization
• Exploiting Unlabeled Data in CNNs by Self-supervised Learning to Rank
• Dilated-Scale-Aware Attention ConvNet For Multi-Class Object Counting

Cited by in corpus (3)

• SwinIR: Image Restoration Using Swin Transformer
• CCTrans: Simplifying and Improving Crowd Counting with Transformer
• Boosting Few-shot Semantic Segmentation with Transformers