论文阅读 [TPAMI-2022] Kernel-Based Density Map Generation for Dense Object Counting

论文阅读 [TPAMI-2022] Kernel-Based Density Map Generation for Dense Object Counting

论文搜索(studyai.com)

搜索论文: Kernel-Based Density Map Generation for Dense Object Counting

搜索论文: http://www.studyai.com/search/whole-site/?q=Kernel-Based+Density+Map+Generation+for+Dense+Object+Counting

关键字(Keywords)

Kernel; Estimation; Feature extraction; Generators; Task analysis; Prediction algorithms; Bandwidth; Crowd counting; vehicle counting; object counting; density map generation; density map estimation; deep learning

机器视觉

人群计数; 物体计数

摘要(Abstract)

Crowd counting is an essential topic in computer vision due to its practical usage in surveillance systems.

由于在监控系统中的实际应用,人群计数是计算机视觉中的一个重要课题。.

The typical design of crowd counting algorithms is divided into two steps.

人群计数算法的典型设计分为两个步骤。.

First, the ground-truth density maps of crowd images are generated from the ground-truth dot maps (density map generation), e.g., by convolving with a Gaussian kernel.

首先,人群图像的地面真实密度图由地面真实点图(密度图生成)生成,例如通过与高斯核卷积。.

Second, deep learning models are designed to predict a density map from an input image (density map estimation).

其次,设计了深度学习模型,从输入图像预测密度图(密度图估计)。.

The density map based counting methods that incorporate density map as the intermediate representation have improved counting performance dramatically.

基于密度图的计数方法结合了密度图作为中间表示,极大地提高了计数性能。.

However, in the sense of end-to-end training, the hand-crafted methods used for generating the density maps may not be optimal for the particular network or dataset used.

然而,在端到端培训的意义上,用于生成密度图的手工方法对于所使用的特定网络或数据集可能不是最优的。.

To address this issue, we propose an adaptive density map generator, which takes the annotation dot map as input, and learns a density map representation for a counter.

为了解决这个问题,我们提出了一种自适应密度图生成器,它将注释点图作为输入,并学习计数器的密度图表示。.

The counter and generator are trained jointly within an end-to-end framework.

计数器和生成器在端到端的框架内进行联合培训。.

We also show that the proposed framework can be applied to general dense object counting tasks.

我们还表明,该框架可以应用于一般的密集目标计数任务。.

Extensive experiments are conducted on 10 datasets for 3 applications: crowd counting, vehicle counting, and general object counting.

在10个数据集上进行了广泛的实验,用于3个应用:人群计数、车辆计数和一般物体计数。.

The experiment results on these datasets confirm the effectiveness of the proposed learnable density map representations…

在这些数据集上的实验结果证实了所提出的可学习密度图表示法的有效性。。.

作者(Authors)

[‘Jia Wan’, ‘Qingzhong Wang’, ‘Antoni B. Chan’]

你可能感兴趣的:(计算机视觉,人工智能,机器学习,CVPR,深度学习)