Abstract
This paper aims to compare the three mainstream solutions for today’s crowd counting and analyze the highlights of each model. In MCNN, they proposed a multi-column parallel convolutional neural network structure that generates population density maps by adapting crowd changes caused by camera view-points and resolution using filters with different size receptive fields. In Switch-CNN, they added a density classifier to the MCNN to enable the use of local density changes in the crowd. In CSRNet, they abandoned the structure of a multi-column convolutional neural network, using the first ten layers of VGG-16 as the front part and the convolutional neural network as the latter part. From the analysis results, CSRNet shows advanced performance. In addition, we analyzed the comparison results of three convolutional neural networks, and derived the trend of convolutional neural network structure.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
