Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

IL-33 Inhibits Hepatitis B Virus through Its Receptor ST2 in Hydrodynamic HBV Mouse Model.

Mediators of inflammation·2020
Same author

The first complete genome sequence and pathogenicity characterization of fowl adenovirus 11 from chickens with inclusion body hepatitis in Pakistan.

Veterinary microbiology·2020
Same author

Inducible MyD88/CD40 synergizes with IL-15 to enhance antitumor efficacy of CAR-NK cells.

Blood advances·2020
Same author

Administration with Vaccinia Virus Encoding Canine Parvovirus 2 <i>vp2</i> Elicits Systemic Immune Responses in Mice and Dogs.

Viral immunology·2020
Same author

[Effect of extracellular vesicles and microRNAs in follicular fluid on follicular development].

Sheng wu gong cheng xue bao = Chinese journal of biotechnology·2020
Same author

Aligned soy protein isolate-modified poly(L-lactic acid) nanofibrous conduits enhanced peripheral nerve regeneration.

Journal of neural engineering·2020

相关实验视频

Updated: Jan 7, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K

使用ST-GCN时空图卷积网络的可穿戴传感器数据驱动的运动姿势识别.

Zhongchen Zhang1,2, Xiaomei Wang3

  • 1College of Physical Education and Health, Yili Normal University, Yining, 835000, China.

Scientific reports
|December 30, 2025
PubMed
概括

这项研究介绍了一种基于可穿戴传感器的动作识别的动态拓适应框架. 它确保了生物机械上可信的连接,即使在非标准的传感器位置上也能达到高精度.

关键词:
动态拓调整的适应.可学习的相邻矩阵.时间空间特征 时间空间特征体育活动的认可体育活动的认可可穿戴式传感器

更多相关视频

Author Spotlight: Enhancing Remote Rehabilitation with Virtual Reality and Electromyography
04:06

Author Spotlight: Enhancing Remote Rehabilitation with Virtual Reality and Electromyography

Published on: January 12, 2024

1.0K
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
07:05

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine

Published on: October 27, 2016

9.6K

相关实验视频

Last Updated: Jan 7, 2026

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K
Author Spotlight: Enhancing Remote Rehabilitation with Virtual Reality and Electromyography
04:06

Author Spotlight: Enhancing Remote Rehabilitation with Virtual Reality and Electromyography

Published on: January 12, 2024

1.0K
Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
07:05

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine

Published on: October 27, 2016

9.6K

科学领域:

  • 生物机械工程 生物机械工程
  • 计算机科学 计算机科学
  • 机器学习 机器学习

背景情况:

  • 空间时间图形卷积网络 (ST-GCNs) 在动作识别方面提供了灵活性.
  • 在ST-GCN中可学习的相邻矩阵面临着在可穿戴系统中非标准的传感器放置的挑战.
  • 保持生物力学上可信的连接对于准确的分析至关重要.

研究的目的:

  • 提出一个动态拓适应的ST-GCN框架,用于使用可穿戴传感器进行强大的动作识别.
  • 为了确保生理上有意义的图形结构,先用人类骨进行初始化.
  • 动态调整拓,以适应传感器位置的变化,而不会影响动态现实性.

主要方法:

  • 开发了一个动态拓适应性ST-GCN框架.
  • 启动可学习的邻近矩阵与人类骨架之前.
  • 员工进行端到端的培训,使用L2规范化和Top-K散散化来进行结构改进和可解释性.

主要成果:

  • 在使用八个IMU传感器的跨用户场景中实现了94.1%的准确性.
  • 在跨设备测试中达到91.5%的准确性.
  • 在非标准化部署条件下,在体育姿势识别方面表现出卓越的稳定性.

结论:

  • 拟议的框架有效地解决了在基于可穿戴设备的动作识别中非标准传感器放置的挑战.
  • 由生物力学先验指导的动态拓适应提高了模型的稳定性和准确性.
  • 该方法确保了生理学上有意义和可解释的图形结构,以进行可靠的人类活动分析.