A-HRNet: Attention Based High Resolution Network for Human pose estimation

被引:13
作者
Li, Ying [1 ]
Wang, Chenxi [2 ]
Cao, Yu [1 ]
Liu, Benyuan [1 ]
Luo, Yan [2 ]
Zhang, Honggang [3 ]
机构
[1] Univ Massachusetts Lowell, Dept Comp Sci, Lowell, MA 01854 USA
[2] Univ Massachusetts Lowell, Dept Elect & Comp Engn, Lowell, MA USA
[3] Univ Massachusetts, Engn Dept, Boston, MA 02125 USA
来源
2020 SECOND INTERNATIONAL CONFERENCE ON TRANSDISCIPLINARY AI (TRANSAI 2020) | 2020年
基金
美国国家科学基金会;
关键词
human pose estimation; attention block;
D O I
10.1109/TransAI49837.2020.00016
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Recently, human pose estimation has received much attention in the research community due to its broad range of application scenarios. Most architectures for human pose estimation use multiple resolution networks, such as Hourglass, CPN, HRNet, etc. High Resolution Network (HRNet) is the latest SOTA architecture improved from Hourglass. In this paper, we propose a novel attention block that leverages a special Channel-Attention branch. We use this attention block as the building block and adopt the architecture of HRNet to build our Attention Based HRNet (A-HRNet). Experiments show that our model can consistently outperform HRNet on different datasets. Moreover, our model achieves the state-of-the-art performance on the COCO keypoint detection val2017 dataset (77.7 AP)(-1).
引用
收藏
页码:75 / 79
页数:5
相关论文
共 50 条
[41]   Multi-resolution Fusion Network for Human Pose Estimation in Low-resolution Images [J].
Kim, Boeun ;
Choo, YeonSeung ;
Jeong, Hea In ;
Kim, Chung-Il ;
Shin, Saim ;
Kim, Jungho .
KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS, 2022, 16 (07) :2328-2344
[42]   Transformer-based rapid human pose estimation network [J].
Wang, Dong ;
Xie, Wenjun ;
Cai, Youcheng ;
Li, Xinjie ;
Liu, Xiaoping .
COMPUTERS & GRAPHICS-UK, 2023, 116 :317-326
[43]   Improving Human Pose Estimation Based on Stacked Hourglass Network [J].
Zou, Xuelian ;
Bi, Xiaojun ;
Yu, Changdong .
NEURAL PROCESSING LETTERS, 2023, 55 (07) :9521-9544
[44]   Improving Human Pose Estimation Based on Stacked Hourglass Network [J].
Xuelian Zou ;
Xiaojun Bi ;
Changdong Yu .
Neural Processing Letters, 2023, 55 :9521-9544
[45]   Human Pose Estimation Model Based on Improved Hourglass Network [J].
Liu Hong ;
Ma Jie ;
Chai Yujing .
LASER & OPTOELECTRONICS PROGRESS, 2021, 58 (20)
[46]   Efficient Dual-Scale Cross-Attention Network for Human Pose Estimation [J].
Wang, Dong ;
Tang, Yiming ;
Cai, Youcheng ;
Liu, Xiaoping .
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2025, 74
[47]   WideHRNet: An Efficient Model for Human Pose Estimation Using Wide Channels in Lightweight High-Resolution Network [J].
Samkari, Esraa ;
Arif, Muhammad ;
AlGhamdi, Manal ;
Al Ghamdi, Mohammed A. .
IEEE ACCESS, 2024, 12 :148990-149000
[48]   Low-resolution human pose estimation [J].
Wang, Chen ;
Zhang, Feng ;
Zhu, Xiatian ;
Ge, Shuzhi Sam .
PATTERN RECOGNITION, 2022, 126
[49]   Lightweight human pose estimation algorithm based on polarized self-attention [J].
Liu, Shengjie ;
He, Ning ;
Wang, Cheng ;
Yu, Haigang ;
Han, Wenjing .
MULTIMEDIA SYSTEMS, 2023, 29 (01) :197-210
[50]   Lightweight human pose estimation algorithm based on polarized self-attention [J].
Shengjie Liu ;
Ning He ;
Cheng Wang ;
Haigang Yu ;
Wenjing Han .
Multimedia Systems, 2023, 29 :197-210