Education, Science, Technology, Innovation and Life
Open Access
Sign In

NF-Net: Crowd Counting Based on Near-Far Network and Dynamic Dual Attention Mechanism

Download as PDF

DOI: 10.23977/jaip.2025.080404 | Downloads: 2 | Views: 42

Author(s)

Zhang Dongming 1, Tian Xueqing 1, Zhao Wenhui 1, Guo Yihan 1, Chen Lijia 1

Affiliation(s)

1 School of Physics and Electronics, Henan University, Kaifeng, 475004, China

Corresponding Author

Tian Xueqing

ABSTRACT

Crowd counting and precise localization in dense scenes are critical tasks in computer vision. Although the point-based prediction framework P2PNet eliminates complex post-processing steps through ensemble prediction, it still faces challenges such as insufficient multi-scale feature extraction and limited ability to perceive key information in complex scenes. To address these issues, this paper improves P2PNet by proposing a model based on a far-near network and a dynamic dual attention mechanism. Specifically, it introduces a far-near adaptive network (FN-Net) and a dynamic dual attention mechanism (DDAM). FN-Net explicitly models continuous scale variations caused by perspective effects by dividing the image into regions based on spatial position and assigning differentiated receptive fields. DDAM focuses on crowded areas through parallel spatial attention and channel attention sub-modules, selects discriminative features, and integrates a dynamic weighted fusion mechanism to adaptively combine the advantages of both attentions. Experiments show that our approach effectively enhances key features while suppressing background noise thus improves crowd counting accuracy.

KEYWORDS

Crowd counting; Point localization; Multi-scale feature modelling; Attention mechanism; Deep learning

CITE THIS PAPER

Zhang Dongming, Tian Xueqing, Zhao Wenhui, Guo Yihan, Chen Lijia, NF-Net: Crowd Counting Based on Near-Far Network and Dynamic Dual Attention Mechanism. Journal of Artificial Intelligence Practice (2025) Vol. 8: 25-31. DOI: http://dx.doi.org/10.23977/jaip.2025.080404.

REFERENCES

[1] Song, Q., Wang, C., Wang, Y., Zhang, Y. and Zhang, C. (2021) Rethinking Counting and Localization in Crowds: A Purely Point-Based Framework. Proceedings of the IEEE/CVF International Conference on Computer Vision, 1, 1-10.
[2] Zhang, Y., Zhou, D., Chen, S., Gao, S. and Ma, Y. (2016) Single-Image Crowd Counting via Multi-Column Convolutional Neural Network. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1, 1-9.
[3] Li, Y., Zhang, X. and Chen, D. (2018) CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1, 1-9.
[4] Liang, D., Chen, X., Xu, W., Zhou, Y. and Bai, X. (2022) An End-to-End Transformer Model for Crowd Localization. European Conference on Computer Vision, 1, 1-16.
[5] Liu, W., Salzmann, M. and Fua, P. (2021) GL: A Generic Framework for Learning Object Counting from Point Supervision. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1, 1-10.
[6] Liang, D., Xu, W., Bai, X. and Zhou, Y. (2022) CCTrans: Simplifying and Improving Crowd Counting with Transformer. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1, 1-10.
[7] Ma, Z., Wei, X., Hong, X. and Gong, Y. (2023) PET: A Purely Point-Based End-to-End Transformer for Crowd Counting. Proceedings of the IEEE/CVF International Conference on Computer Vision, 1, 1-10.
[8] Zhang, L., Li, H., Chen, X. and Wang, Y. (2024) HSNet: Hierarchical Scale Network for Crowd Counting. Engineering Applications of Artificial Intelligence, 128, 107589.
[9] Wang, Y., Zhang, J., Zhang, Y., Wang, L. and Wang, C. (2023) DDC: A Dual-branch Dilated Convolutional Network for Crowd Counting. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1, 1-10.
[10] Wang, Q., Li, J., Zhang, Y., Zhou, Y. and Yang, J. (2024) M2PLNet: Multi-scale Multi-patch Learning Network for Crowd Counting. IEEE International Conference on Multimedia and Expo, 1, 1-6.
[11] Wang, Z., Liu, Y., Chen, Z., Li, M. and Zhang, T. (2025) EHNet: Enhancing Crowd Counting with A Scale-Aware Hierarchical Network. arXiv preprint, arXiv:2503.12061, 1-10.

Downloads: 16470
Visits: 571364

Sponsors, Associates, and Links


All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.