Abstract
Crowd counting, an essential technology with numerous applications, often encounters challenges due to non-uniform crowd distributions and noisy backgrounds in congested scenes. To address these issues, this paper proposes the utilization of depth information as an independent indicator. Specifically, we introduce a depth-guided attention strategy (DAS) to fuse depth and crowd density information, effectively modeling the relationship between crowd density and depth of field. Additionally, we propose a depth-guided method to generate the target density map by leveraging the negative correlation between the depth of field and head size in crowd scenes, enabling better supervised learning. To achieve fast inference speed, we design two lightweight crowd counting networks within a knowledge distillation framework that require only a small number of parameters. Furthermore, we propose a two-step network inference algorithm to reduce counting errors. Extensive experiments conducted on four challenging datasets demonstrate that our proposed methods significantly improve counting accuracy over baseline networks.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Abousamra, S., Hoai, M., Samaras, D., Chen, C.: Localization in the crowd with topological constraints. In: AAAI Conference on Artificial Intelligence, vol. 35, pp. 872–881 (2021)
Cao, X., Wang, Z., Zhao, Y., Su, F.: Scale aggregation network for accurate and efficient crowd counting. In: European Conference on Computer Vision, pp. 734–750 (2018)
Han, K., Wang, Y., Tian, Q., Guo, J., Xu, C., Xu, C.: Ghostnet: more features from cheap operations. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 1580–1589 (2020)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778 (2016)
Howard, A.G., et al.: Mobilenets: efficient convolutional neural networks for mobile vision applications. ar**v preprint ar**v:1704.04861 (2017)
Hua, C., Xu, K., Tong, T.: Crowd counting with dilated inception convolution. In: International Conference on Computing and Artificial Intelligence, pp. 208–215 (2021)
Idrees, H., Saleemi, I., Seibert, C., Shah, M.: Multi-source multi-scale counting in extremely dense crowd images. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 2547–2554 (2013)
Idrees, H., et al.: Composition loss for counting, density map estimation and localization in dense crowds. In: European Conference on Computer Vision, pp. 532–546 (2018)
Jiang, G., Wu, R., Huo, Z., Zhao, C., Luo, J.: Ligmsaet: lightweight multi-scale adaptive convolutional neural network for dense crowd counting. Expert Syst. Appl. 197, 116662 (2022)
Li, Y., Zhang, X., Chen, D.: Csrnet: dilated convolutional neural networks for understanding the highly congested scenes. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 1091–1100 (2018)
Li, Z., Lu, S., Dong, Y., Guo, J.: Msffa: a multi-scale feature fusion and attention mechanism network for crowd counting. Visual Comput. 39(3), 1045–1056 (2023)
Lian, D., Li, J., Zheng, J., Luo, W., Gao, S.: Density map regression guided detection network for rgb-d crowd counting and localization. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 1821–1830 (2019)
Liu, L., Chen, J., Wu, H., Chen, T., Li, G., Lin, L.: Efficient crowd counting via structured knowledge transfer. In: ACM International Conference on Multimedia, pp. 2645–2654 (2020)
Liu, W., Salzmann, M., Fua, P.: Context-aware crowd counting. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 5099–5108 (2019)
Ma, X., Du, S., Liu, Y.: A lightweight neural network for crowd analysis of images with congested scenes. In: IEEE International Conference on Image Processing, pp. 979–983 (2019)
Ma, Z., Wei, X., Hong, X., Lin, H., Qiu, Y., Gong, Y.: Learning to count via unbalanced optimal transport. In: AAAI Conference on Artificial Intelligence, vol. 35, pp. 2319–2327 (2021)
Oh, M.h., Olsen, P., Ramamurthy, K.N.: Crowd counting with decomposed uncertainty. In: AAAI Conference on Artificial Intelligence, vol. 34, pp. 11799–11806 (2020)
Ranftl, R., Lasinger, K., Hafner, D., Schindler, K., Koltun, V.: Towards robust monocular depth estimation: mixing datasets for zero-shot cross-dataset transfer. IEEE Trans. Pattern Anal. Mach. Intell. 44, 1623–1637 (2020)
Shu, W., Wan, J., Tan, K.C., Kwong, S., Chan, A.B.: Crowd counting in the frequency domain. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 19618–19627 (2022)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. ar**v preprint ar**v:1409.1556 (2014)
Wan, J., Liu, Z., Chan, A.B.: A generalized loss function for crowd counting and localization. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 1974–1983 (2021)
Wang, B., Liu, H., Samaras, D., Nguyen, M.H.: Distribution matching for crowd counting. Adv. Neural Inf. Process. Syst. 33, 1595–1607 (2020)
Wang, M., Cai, H., Dai, Y., Gong, M.: Dynamic mixture of counter network for location-agnostic crowd counting. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 167–177 (2023)
Xu, C., et al.: Autoscale: learning to scale for crowd counting. Int. J. Comput. Vision 130(2), 405–434 (2022)
Yang, S.D., Su, H.T., Hsu, W.H., Chen, W.C.: Deccnet: depth enhanced crowd counting. In: IEEE International Conference on Computer Vision Workshops, p. 4521–4530 (2019)
, Zhai, W., Gao, M., Li, Q., Jeon, G., Anisetti, M.: Fpanet: feature pyramid attention network for crowd counting. Appl. Intell. 1–18 (2023)
Zhang, Y., Zhou, D., Chen, S., Gao, S., Ma, Y.: Single-image crowd counting via multi-column convolutional neural network. In: IEEE Conference on Computer Vision and Pattern Recognition, pp. 589–597 (2016)
Zhou, F., et al.: Comal: compositional multi-scale feature enhanced learning for crowd counting. Multimedia Tools Appl., 1–20 (2022)
Zhu, F., Yan, H., Chen, X., Li, T.: Real-time crowd counting via lightweight scale-aware network. Neurocomputing 472, 54–67 (2022)
Acknowledgements
This work was financially supported by the Guangdong Basic and Applied Basic Research Foundation (No. 2022A1515010119) and the National Natural Science Foundation of China (No. 62071201, No. U2031104).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Chen, H., Li, Z., Bhanu, B., Lu, D., Han, X. (2023). A Depth-Guided Attention Strategy for Crowd Counting. In: Iliadis, L., Papaleonidas, A., Angelov, P., Jayne, C. (eds) Artificial Neural Networks and Machine Learning – ICANN 2023. ICANN 2023. Lecture Notes in Computer Science, vol 14263. Springer, Cham. https://doi.org/10.1007/978-3-031-44204-9_3
Download citation
DOI: https://doi.org/10.1007/978-3-031-44204-9_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44203-2
Online ISBN: 978-3-031-44204-9
eBook Packages: Computer ScienceComputer Science (R0)