CAGNet: coordinated attention guidance network for RGB-T crowd counting
Estimating crowd density is a demanding task that has garnered significant research attention in urban planning, intelligent transportation, and other related fields. This study utilizes RGB and thermal images to leverage multimodal information and introduces a coordinated attention guidance network...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/180135 |
| _version_ | 1826117239332405248 |
|---|---|
| author | Yang, Xun Zhou, Wujie Yan, Weiqing Qian, Xiaohong |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Yang, Xun Zhou, Wujie Yan, Weiqing Qian, Xiaohong |
| author_sort | Yang, Xun |
| collection | NTU |
| description | Estimating crowd density is a demanding task that has garnered significant research attention in urban planning, intelligent transportation, and other related fields. This study utilizes RGB and thermal images to leverage multimodal information and introduces a coordinated attention guidance network (CAGNet) for RGB-thermal (RGB-T) crowd counting. The framework enhances the expressive capabilities of backbone features by incorporating overall and local relationships through the information enhancement unit module, which utilizes the context coordination perception module for horizontal information mining, interactively compensating for the continuity of spatial information. Subsequently, it utilizes diverse multi-level information features for hierarchical intersection and progression, resulting in an accurate crowd counting density map. Experimental results on the RGBT-CC benchmark dataset demonstrate the robustness and effectiveness of CAGNet for RGB-T crowd counting. Furthermore, the proposed CAGNet can be extended to crowd density estimation and has achieved high performance on the ShanghaiTechRGBD and DroneRGBT datasets. The former dataset comprises paired RGB images and depth maps. The code and model for this research are available at https://github.com/WBangG/CAGNet. |
| first_indexed | 2024-10-01T04:24:05Z |
| format | Journal Article |
| id | ntu-10356/180135 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:24:05Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1801352024-09-18T06:13:54Z CAGNet: coordinated attention guidance network for RGB-T crowd counting Yang, Xun Zhou, Wujie Yan, Weiqing Qian, Xiaohong School of Computer Science and Engineering Computer and Information Science Urban planning Intelligent transportation Estimating crowd density is a demanding task that has garnered significant research attention in urban planning, intelligent transportation, and other related fields. This study utilizes RGB and thermal images to leverage multimodal information and introduces a coordinated attention guidance network (CAGNet) for RGB-thermal (RGB-T) crowd counting. The framework enhances the expressive capabilities of backbone features by incorporating overall and local relationships through the information enhancement unit module, which utilizes the context coordination perception module for horizontal information mining, interactively compensating for the continuity of spatial information. Subsequently, it utilizes diverse multi-level information features for hierarchical intersection and progression, resulting in an accurate crowd counting density map. Experimental results on the RGBT-CC benchmark dataset demonstrate the robustness and effectiveness of CAGNet for RGB-T crowd counting. Furthermore, the proposed CAGNet can be extended to crowd density estimation and has achieved high performance on the ShanghaiTechRGBD and DroneRGBT datasets. The former dataset comprises paired RGB images and depth maps. The code and model for this research are available at https://github.com/WBangG/CAGNet. This work was supported by the National Natural Science Foundation of China (61502429); and the Zhejiang Provincial Natural Science Foundation of China (LY18F020012). 2024-09-18T06:13:54Z 2024-09-18T06:13:54Z 2024 Journal Article Yang, X., Zhou, W., Yan, W. & Qian, X. (2024). CAGNet: coordinated attention guidance network for RGB-T crowd counting. Expert Systems With Applications, 243, 122753-. https://dx.doi.org/10.1016/j.eswa.2023.122753 0957-4174 https://hdl.handle.net/10356/180135 10.1016/j.eswa.2023.122753 2-s2.0-85183588558 243 122753 en Expert Systems with Applications © 2023 Elsevier Ltd. All rights reserved. |
| spellingShingle | Computer and Information Science Urban planning Intelligent transportation Yang, Xun Zhou, Wujie Yan, Weiqing Qian, Xiaohong CAGNet: coordinated attention guidance network for RGB-T crowd counting |
| title | CAGNet: coordinated attention guidance network for RGB-T crowd counting |
| title_full | CAGNet: coordinated attention guidance network for RGB-T crowd counting |
| title_fullStr | CAGNet: coordinated attention guidance network for RGB-T crowd counting |
| title_full_unstemmed | CAGNet: coordinated attention guidance network for RGB-T crowd counting |
| title_short | CAGNet: coordinated attention guidance network for RGB-T crowd counting |
| title_sort | cagnet coordinated attention guidance network for rgb t crowd counting |
| topic | Computer and Information Science Urban planning Intelligent transportation |
| url | https://hdl.handle.net/10356/180135 |
| work_keys_str_mv | AT yangxun cagnetcoordinatedattentionguidancenetworkforrgbtcrowdcounting AT zhouwujie cagnetcoordinatedattentionguidancenetworkforrgbtcrowdcounting AT yanweiqing cagnetcoordinatedattentionguidancenetworkforrgbtcrowdcounting AT qianxiaohong cagnetcoordinatedattentionguidancenetworkforrgbtcrowdcounting |