期刊
IEEE TRANSACTIONS ON IMAGE PROCESSING
卷 29, 期 -, 页码 2714-2727出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIP.2019.2952083
关键词
Crowd counting; density level analysis; pan-density evaluation; convolutional neural networks
资金
- National Key R&D Program of China [2018YFB1305104]
- National Natural Science Foundation of China [NSFC 61673118]
- Shanghai Municipal Science and Technology Major Project [2018SHZDZX01]
- ZJLab
- Pennsylvania State University
Crowd counting is a highly challenging problem in computer vision and machine learning. Most previous methods have focused on consistent density crowds, i.e., either a sparse or a dense crowd, meaning they performed well in global estimation while neglecting local accuracy. To make crowd counting more useful in the real world, we propose a new perspective, named pan-density crowd counting, which aims to count people in varying density crowds. Specifically, we propose the Pan-Density Network (PaDNet) which is composed of the following critical components. First, the Density-Aware Network (DAN) contains multiple subnetworks pretrained on scenarios with different densities. This module is capable of capturing pan-density information. Second, the Feature Enhancement Layer (FEL) effectively captures the global and local contextual features and generates a weight for each density-specific feature. Third, the Feature Fusion Network (FFN) embeds spatial context and fuses these density-specific features. Further, the metrics Patch MAE (PMAE) and Patch RMSE (PRMSE) are proposed to better evaluate the performance on the global and local estimations. Extensive experiments on four crowd counting benchmark datasets, the ShanghaiTech, the UCFCC50, the UCSD, and the UCF-QNRF, indicate that PaDNet achieves state-of-the-art recognition performance and high robustness in pan-density crowd counting.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据