期刊
IEEE TRANSACTIONS ON MULTIMEDIA
卷 23, 期 -, 页码 1060-1068出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TMM.2020.2992979
关键词
Convolution; Estimation; Feature extraction; Loss measurement; Image segmentation; Computer architecture; Measurement uncertainty; Crowd counting; density estimation; detection
资金
- National Natural Science Fund of China [61725501, 61976037]
- Specialized Research Fund for the Doctoral Program of Higher Education of China [20121102110032]
This paper introduces a network called DENet, consisting of a detection network and an encoder-decoder estimation network, for counting people or objects with varying scales and densities. Evaluation on three datasets shows that DENet achieves lower Mean Absolute Error (MAE) than state-of-the-art methods.
Counting people or objects with significantly varying scales and densities has attracted much interest from the research community and yet it remains an open problem. In this paper, we propose a simple but efficient and effective network, named DENet, which is composed of two components, i.e., a detection network (DNet) and an encoder-decoder estimation network (ENet). We first run the DNet on the input image to detect and count individuals who can be segmented clearly. Then, the ENet is utilized to estimate the density maps of the remaining areas, typically with low resolution and high densities where individuals cannot be detected. For this purpose, we propose a modified Xception network as the encoder for feature extraction and a combination of dilated convolution and transposed convolution as the decoder. When evaluated on the ShanghaiTech Part A, UCF and WorldExpo'10 datasets, our DENet has achieved lower Mean Absolute Error (MAE) than those of the state-of-the-art methods.
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