In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation *** this paper,we aim to reduce the annotation cost of crowd datasets,a...
详细信息
In a crowd density estimation dataset,the annotation of crowd locations is an extremely laborious task,and they are not taken into the evaluation *** this paper,we aim to reduce the annotation cost of crowd datasets,and propose a crowd density estimation method based on weakly-supervised learning,in the absence of crowd position supervision information,which directly reduces the number of crowds by using the number of pedestrians in the image as the supervised *** this purpose,we design a new training method,which exploits the correlation between global and local image features by incremental learning to train the ***,we design a parent-child network(PC-Net)focusing on the global and local image respectively,and propose a linear feature calibration structure to train the PC-Net simultaneously,and the child network learns feature transfer factors and feature bias weights,and uses the transfer factors and bias weights to linearly feature calibrate the features extracted from the Parent network,to improve the convergence of the network by using local features hidden in the crowd *** addition,we use the pyramid vision transformer as the backbone of the PC-Net to extract crowd features at different levels,and design a global-local feature loss function(L2).We combine it with a crowd counting loss(LC)to enhance the sensitivity of the network to crowd features during the training process,which effectively improves the accuracy of crowd density *** experimental results show that the PC-Net significantly reduces the gap between fullysupervised and weakly-supervised crowd density estimation,and outperforms the comparison methods on five datasets of Shanghai Tech Part A,ShanghaiTech Part B,UCF_CC_50,UCF_QNRF and JHU-CROWD++.
This research investigates the efficacy of various machine learning algorithms in detecting malware. Utilizing a dataset of 20,000 observations with 33 features,28 classification algorithms are applied and evaluated t...
详细信息
Agriculture is one of the most important occupations for the majority of people in the world's second largest populated country, India. However, due to a lack of education, accurate information, and India's ra...
详细信息
For better patient diagnosis and treatment, medical facilities need to be advanced. With the assistance of machine learning, we can large and sophisticated medical datasets for analyzing them and getting clinical insi...
详细信息
In the previous several decades, there have been tremendous improvements in the analysis of human movement in the field of sports biomechanics and rehabilitation. Driven by the demand for faster and more complex ways ...
详细信息
Written applications, with standardized fields like date, subject, sender and receiver, serve various purposes such as leave requests, misconduct appeals and suggestions. Surprisingly, even in today’s tech-driven wor...
详细信息
There have been tremendous advancements in technology that have led to an increase in the needs of people. This, in turn, has led to more loan approval requests in the banking sector. Some attributes are considered to...
详细信息
The area lies near Yamuna River in Delhi faces a frequent flood event, and with the increasing frequency and intensity of flood events highly impacted the human lives, infrastructure, and agriculture. Hence, there is ...
详细信息
Nowadays we are seeing that the merchandise (product) we are buying can have (20-25) % chances that it can be fictitious merchandise. And this also affects the sales of the company. This paper discusses an application...
详细信息
Virtualization is a technology that allows us to create numerous simulated environments, which means instead of using an actual version of something a virtual copy is created. It makes things easier to manage. It was ...
详细信息
暂无评论