Joint graph convolution networks and transformer for human pose estimation in sports technique analysis

被引:6
|
作者
Cheng, Hongren [1 ,2 ]
Wang, Jing [3 ]
Zhao, Anran [4 ]
Zhong, Yaping [1 ,2 ]
Li, Jingli [5 ]
Dong, Liangshan [6 ]
机构
[1] Wuhan Sports Univ, Sports Big Data Res Ctr, Wuhan 430079, Peoples R China
[2] Hubei Prov Sports & Hlth Innovat Dev Res Ctr, Wuhan 430079, Hubei, Peoples R China
[3] Chongqing Univ Posts & Telecommun, Sch Automat, Chongqing 400065, Peoples R China
[4] Wuhan Univ, Sch Remote Sensing & Informat Engn, Wuhan 430079, Peoples R China
[5] Huazhong Univ Sci & Technol, Sch Phys Educ, Wuhan 430074, Peoples R China
[6] China Univ Geosci, Sch Phys Educ, Wuhan 430074, Peoples R China
关键词
Human pose estimation; Graph convolutional network; Transformer; The topological structure between; IMAGE STEGANOGRAPHY METHOD;
D O I
10.1016/j.jksuci.2023.101819
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Human pose estimation has various applications in domains such as sports technology analysis, virtual reality, and education. However, most previous studies focused on the respective feature representations of keypoints, but disregarded the topological relationship among keypoints. To address this challenge, we propose GTPose, a network structure that integrates graph convolutional networks and Transform. First of all, a set of multi-scale convolution operations are applied to extract local feature maps of images. Secondly, the positions of keypoints are roughly estimated by using Transform to process the sequential relations between feature maps. Finally, GCN is adopted to model the topological structure between keypoints to accurately locate the location of keypoints and learn feature representations. The performance of GTPose is evaluated on two real datasets: MS COCO and MPII. Experimental results demonstrate that GTPose outperforms other methods in human pose estimation tasks. In addition, experimental results also show that the spatial relationship between keypoints is effective for accurately characterizing keypoints.
引用
收藏
页数:8
相关论文
共 50 条
  • [21] Dual Graph Networks for Pose Estimation in Crowded Scenes
    Tu, Jun
    Wu, Gangshan
    Wang, Limin
    INTERNATIONAL JOURNAL OF COMPUTER VISION, 2024, 132 (03) : 633 - 653
  • [22] Dual Graph Networks for Pose Estimation in Crowded Scenes
    Jun Tu
    Gangshan Wu
    Limin Wang
    International Journal of Computer Vision, 2024, 132 (3) : 633 - 653
  • [23] Graph transformer based dynamic multiple graph convolution networks for traffic flow forecasting
    Hu, Yongli
    Peng, Ting
    Guo, Kan
    Sun, Yanfeng
    Gao, Junbin
    Yin, Baocai
    IET INTELLIGENT TRANSPORT SYSTEMS, 2023, 17 (09) : 1835 - 1845
  • [24] GHand: A Graph Convolution Network for 3D Hand Pose Estimation
    Wang, Pengsheng
    Xue, Guangtao
    Li, Pin
    Kim, Jinman
    Sheng, Bin
    Mao, Lijuan
    ADVANCES IN COMPUTER GRAPHICS, CGI 2020, 2020, 12221 : 374 - 381
  • [25] Deep Semantic Graph Transformer for Multi-View 3D Human Pose Estimation
    Zhang, Lijun
    Zhou, Kangkang
    Lu, Feng
    Zhou, Xiang-Dong
    Shi, Yu
    THIRTY-EIGHTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 38 NO 7, 2024, : 7205 - 7214
  • [26] Towards infrared human pose estimation via Transformer
    Zhu, Zhilei
    Dong, Wanli
    Gao, Xiaoming
    Peng, Anjie
    2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, 2023,
  • [27] Multibranch Attention Graph Convolutional Networks for 3-D Human Pose Estimation
    Yin, Yanfang
    Liu, Ming
    Zhu, Qigang
    Zhang, Shuaishuai
    Hussien, Naseer Ali
    Fan, Yong
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2023, 72
  • [28] Human Pose Estimation and Object Interaction for Sports Behaviour
    Arif, Ayesha
    Ghadi, Yazeed Yasin
    Alarfaj, Mohammed
    Jalal, Ahmad
    Kamal, Shaharyar
    Kim, Dong-Seong
    CMC-COMPUTERS MATERIALS & CONTINUA, 2022, 72 (01): : 1 - 18
  • [29] 3D Human Pose Estimation Using Mobius Graph Convolutional Networks
    Azizi, Niloofar
    Possegger, Horst
    Rodola, Emanuele
    Bischof, Horst
    COMPUTER VISION - ECCV 2022, PT I, 2022, 13661 : 160 - 178
  • [30] 3D human pose estimation with multi-scale graph convolution and hierarchical body pooling
    Ke Huang
    TianQi Sui
    Hong Wu
    Multimedia Systems, 2022, 28 : 403 - 412