DEO-Net: joint density estimation and object detection for crowd counting
Automated crowd counting has emerged as a vision-based measurement method for crowd analysis and management. However, current methods based on density maps still suffer from challenges related to background noise and blurring effects. To address the limitations, this work proposes a deep neural netw...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/180575 |
| _version_ | 1826121071299919872 |
|---|---|
| author | Phan, Duc Tri Gao, Jianjun Lu, Ye Yap, Kim-Hui Garg, Kratika Han, Boon Siew |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Phan, Duc Tri Gao, Jianjun Lu, Ye Yap, Kim-Hui Garg, Kratika Han, Boon Siew |
| author_sort | Phan, Duc Tri |
| collection | NTU |
| description | Automated crowd counting has emerged as a vision-based measurement method for crowd analysis and management. However, current methods based on density maps still suffer from challenges related to background noise and blurring effects. To address the limitations, this work proposes a deep neural network, named joint density estimation and object detection (DEO-Net), specifically designed to generate high-quality density estimation maps. DEO-Net bridges the gap between detection and density estimation-based methods in crowd counting. The key contributions of this research are as follows: 1) DEO-Net incorporates object detection for more accurate crowd localization; 2) the network training is optimized with an independent structural similarity index (I-SSIM) and curriculum losses to better learn local structural information and recognize local maxima; and 3) the experimental results demonstrate the state-of-the-art (SOTA) performance of the proposed DEO-Net with mean absolute error (MAE) values of 54.2, 6.2, 83.1, and 57.3 on the ShangHaiTechA, ShanghaiTechB, UCF_QNRF, and JHU-CROWD++ public datasets, respectively. |
| first_indexed | 2025-03-09T12:52:44Z |
| format | Journal Article |
| id | ntu-10356/180575 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2025-03-09T12:52:44Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1805752024-10-18T15:41:49Z DEO-Net: joint density estimation and object detection for crowd counting Phan, Duc Tri Gao, Jianjun Lu, Ye Yap, Kim-Hui Garg, Kratika Han, Boon Siew School of Electrical and Electronic Engineering Engineering Background noise Crowd counting Automated crowd counting has emerged as a vision-based measurement method for crowd analysis and management. However, current methods based on density maps still suffer from challenges related to background noise and blurring effects. To address the limitations, this work proposes a deep neural network, named joint density estimation and object detection (DEO-Net), specifically designed to generate high-quality density estimation maps. DEO-Net bridges the gap between detection and density estimation-based methods in crowd counting. The key contributions of this research are as follows: 1) DEO-Net incorporates object detection for more accurate crowd localization; 2) the network training is optimized with an independent structural similarity index (I-SSIM) and curriculum losses to better learn local structural information and recognize local maxima; and 3) the experimental results demonstrate the state-of-the-art (SOTA) performance of the proposed DEO-Net with mean absolute error (MAE) values of 54.2, 6.2, 83.1, and 57.3 on the ShangHaiTechA, ShanghaiTechB, UCF_QNRF, and JHU-CROWD++ public datasets, respectively. Agency for Science, Technology and Research (A*STAR) Submitted/Accepted version This research is supported by the Agency for Science, Technology, and Research (A∗STAR) under its IAF-ICP Programme I2001E0067 and the Schaeffler Hub for Advanced Research at NTU. 2024-10-14T01:50:40Z 2024-10-14T01:50:40Z 2024 Journal Article Phan, D. T., Gao, J., Lu, Y., Yap, K., Garg, K. & Han, B. S. (2024). DEO-Net: joint density estimation and object detection for crowd counting. IEEE Transactions On Instrumentation and Measurement, 73, 5027911-. https://dx.doi.org/10.1109/TIM.2024.3441018 0018-9456 https://hdl.handle.net/10356/180575 10.1109/TIM.2024.3441018 2-s2.0-85200807467 73 5027911 en I2001E0067 IEEE Transactions on Instrumentation and Measurement © 2024 IEEE. All rights reserved. This article may be downloaded for personal use only. Any other use requires prior permission of the copyright holder. The Version of Record is available online at http://doi.org/10.1109/TIM.2024.3441018. application/pdf |
| spellingShingle | Engineering Background noise Crowd counting Phan, Duc Tri Gao, Jianjun Lu, Ye Yap, Kim-Hui Garg, Kratika Han, Boon Siew DEO-Net: joint density estimation and object detection for crowd counting |
| title | DEO-Net: joint density estimation and object detection for crowd counting |
| title_full | DEO-Net: joint density estimation and object detection for crowd counting |
| title_fullStr | DEO-Net: joint density estimation and object detection for crowd counting |
| title_full_unstemmed | DEO-Net: joint density estimation and object detection for crowd counting |
| title_short | DEO-Net: joint density estimation and object detection for crowd counting |
| title_sort | deo net joint density estimation and object detection for crowd counting |
| topic | Engineering Background noise Crowd counting |
| url | https://hdl.handle.net/10356/180575 |
| work_keys_str_mv | AT phanductri deonetjointdensityestimationandobjectdetectionforcrowdcounting AT gaojianjun deonetjointdensityestimationandobjectdetectionforcrowdcounting AT luye deonetjointdensityestimationandobjectdetectionforcrowdcounting AT yapkimhui deonetjointdensityestimationandobjectdetectionforcrowdcounting AT gargkratika deonetjointdensityestimationandobjectdetectionforcrowdcounting AT hanboonsiew deonetjointdensityestimationandobjectdetectionforcrowdcounting |