TANG Zhixuan,GAO Yuxiang.Lightweight Human Pose Estimation Algorithm based on Improved Dual Attention Mechanism[J].Journal of Chengdu University of Information Technology,2025,40(02):157-162.[doi:10.16836/j.cnki.jcuit.2025.02.006]
基于改进双注意力机制的轻量型人体姿态检测算法
- Title:
- Lightweight Human Pose Estimation Algorithm based on Improved Dual Attention Mechanism
- 文章编号:
- 2096-1618(2025)02-0157-06
- Keywords:
- human pose estimation; high-resolution network; attention mechanism; feature reuse; lightweight network
- 分类号:
- TP391
- 文献标志码:
- A
- 摘要:
- 为提高轻量化人体姿态检测算法的准确率,提出一种新的双注意力机制的轻量化高分辨率人体姿态检测网络ES-LHRNet。借鉴HRNet(high-resolution network)的框架,采用稠密连接网络和堆叠的轻量化倒残差结构进行特征提取,并提出一种新的融合通道注意力和空间注意力机制的双注意力机制捕获位置信息和通道信息提升算法准确率。相比于HRNet,ES-LHRNet参数量减少86%,运算复杂度下降73%,并且提出的ESAM使在数据集MS COCO 2017上的检测结果平均精度提升了1.6个百分点。
- Abstract:
- To improve the accuracy of the lightweight human posture detection algorithm,a new dual-attention mechanism of lightweight high-resolution human pose estimation network ES-LHRNet was proposed.Based on the overall framework of HRNet(high-resolution network)as a reference,this paper carried out feature extraction by Densenet and stacked lightweight inverse residual structure,and a new dual attention mechanism that integrates spatial attention mechanism and channel attention mechanism to capture location information and channel information promotion algorithm.Compared with HRNet,the number of parameters in this model is reduced by 86%,the computational complexity is reduced by 73%,and the proposed ESAM improves the map of detection results on dataset MS COCO 2017 by 1.6 percentage.
参考文献/References:
[1] ToshevA,Szegedy C.Deeppose:Human pose estimation via deep neural networks[C].Proceedings of the IEEE conference on computer vision and pattern recognition.2014:1653-1660.
[2] WeiS E,Ramakrishna V,Kanade T,et al.Convolutional pose machines[C].Proceedings of the IEEE conference on Computer Vision and Pattern Recognition.2016:4724-4732.
[3] Cao Z,SimonT,Wei S E,et al.Realtime multi-person 2d pose estimation using part affinity fields[C].Proceedings of the IEEE conference on computer vision and pattern recognition.2017:7291-7299.
[4] Chen Y,Wang Z,Peng Y,et al.Cascaded pyramid network for multi-person pose estimation[C].Proceedings of the IEEE conference on computer vision and pattern recognition.2018:7103-7112.
[5] Sun K,Xiao B,Liu D,et al.Deep high-resolution representation learning for human pose estimation[C].Proceedings of the IEEE/CVF conference on computer vision and pattern recognition.2019:5693-5703.
[6] Yu C,Xiao B,Gao C,et al.Lite-hrnet:A lightweight high-resolution network[C].Proceedings of the IEEE/CVF conference on computer vision and pattern recognition.2021:10440-10450.
[7] Ma N,Zhang X,Zheng H T,et al.Shufflenet v2:Practical guidelines for efficient cnn architecture design[C].Proceedings of the European conference on computer vision(ECCV).2018:116-131.
[8] 卢官明,卢峻禾,陈晨.基于深度学习的二维人体姿态估计研究进展[J].南京邮电大学学报(自然科学版),2024,44(1):44-55.
[9] He K,Zhang X,Ren S,et al.Deep residual learning for image recognition[C].Proceedings of the IEEE conference on computer vision and pattern recognition.2016:770-778.
[10] Huang G,Liu Z,Van Der Maaten L,et al.Densely connected convolutional networks[C].Proceedings of the IEEE conference on computer vision and pattern recognition.2017:4700-4708.
[11] Woo S,Park J,Lee J Y,et al.Cbam:Convolutional block attention module[C].Proceedings of the European conference on computer vision(ECCV),2018:3-19.
[12] Wang Q,Wu B,Zhu P,et al.ECA-Net:Efficient channel attention for deep convolutional neural networks[C].Proceedings of the IEEE/CVF conference on computer vision and pattern recognition,2020:11534-11542.
[13] 李丽,张荣芬,刘宇红,等.基于多尺度注意力机制的高分辨率网络人体姿态估计[J].计算机应用研究,2022,39(11):3487-3491.
[14] 朱翠涛,李博.基于高分辨率网络的人体姿态估计[J].中南民族大学学报(自然科学版),2023,42(2):229-237.
备注/Memo
收稿日期:2023-09-20
通信作者:高瑜翔.E-mail:gaoyuxiang@cuit.edu.cn