Abstract
Crowd counting is a challenging task dealing with the variation of an object scale and a crowd density. Existing works have emphasized on skip connections by integrating shallower layers with deeper layers, where each layer extracts features in a different object scale and crowd density. However, only high-level features are emphasized while ignoring low-level features. This paper proposes an estimation network by passing high-level features to shallow layers and emphasizing its low-level feature. Since an estimation network is a hierarchical network, a high-level feature is also emphasized by an improved low-level feature. Our estimation network consists of two identical networks for extracting a high-level feature and estimating the final result. To preserve semantic information, dilated convolution is employed without resizing the feature map. Our method was tested in three datasets for counting humans and vehicles in a crowd image. The counting performance is evaluated by mean absolute error and root mean squared error indicating the accuracy and robustness of an estimation network, respectively. The experimental result shows that our network outperforms other related works in a high crowd density and is effective for reducing over-counting error in the overall case.
Highlights
In recent years, crowd counting has an important role in a wide range of applications
An actual density map derived from Equation (2) was calculated as the ground truth of an estimation network
These connections help to improve the quality of feature maps in shallow layers by inserting the high-level features from deeper layers
Summary
Crowd counting has an important role in a wide range of applications (e.g., video surveillance, traffic monitoring [1], public safety, urban planning [2], and so on). To solve the scaling problem, estimation networks from previous studies have trended to implement with several image resolutions and convolutional layers (Conv layers) for handling various object scales Their valuable information could be lost by resizing crowd images and their results were limited by the network configuration in different datasets. Skip connections have been added to the estimation network for reusing feature maps, rather than building other networks This technique is called ‘skip-network’ which is effective only on objects with large scales because high-level feature maps are only emphasized. The proposed network utilized the backward connection for passing high-level features from deeper layers to shallower layers and reducing false positives in crowd counting. A general flowchart of crowd counting using deep learning-based technique
Sign-in/Register to view highlights & summary 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.
