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

Physical activity and health-related quality of life in Korean postmenopausal women with and without chronic diseases: a population-based study using the KNHANES 2016-2020 database.

Menopause (New York, N.Y.)·2026
Same author

Prediction Model for Low Bone Mineral Density in Cancer Survivors and Age-Matched Controls Using a Causal Bayesian Network: A Nationwide Population-Based Study in Korea.

Clinical Medicine Insights. Oncology·2026
Same author

Association of Multiple-trait Polygenic Risk Score with Obesity and Cardiometabolic Diseases in Korean Population.

Genomics, proteomics & bioinformatics·2025
Same author

Effects of NSAIDs on Early CKD Development: A 10-Year Population-Based Study Using the Korean Senior Cohort.

Drugs & aging·2025
Same author

Associations of dietary patterns and lower urinary tract symptoms (LUTS) in Korean adults.

Nutrition research and practice·2025
Same author

Anthropometric Changes in Female Participants Enrolled in a Combined Aerobic and Resistance Training Program for Longer than 1 Year: A Cohort Study.

Journal of obesity & metabolic syndrome·2024

相关实验视频

Updated: Jun 15, 2025

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
10:23

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans

Published on: September 8, 2023

2.7K

通过图形的3D人体姿势估计识别向前的头部姿势 卷积网络:开发和可行性研究

Haedeun Lee1, Bumjo Oh2,3, Seung-Chan Kim1

  • 1Machine Learning Systems Laboratory, School of Sports Science, Sungkyunkwan University, Suwon, Gyunggi-do, Republic of Korea.

JMIR formative research
|August 26, 2024
PubMed
概括

前向头部姿势 (FHP) 可能会导致疼痛和疲劳. 一个新的系统使用3D姿势估计和图形卷积网络 (GCN) 来从2D图像中检测FHP,从而实现姿势校正.

关键词:
算法算法是一种算法.深度学习是一种深度学习.前进的头部姿势 前进的头部姿势图表是指图表中的图形.图表 卷积网络 卷积网络图表神经网络的神经网络人类姿势估计估计伤害预测 伤害预测机器学习是机器学习.神经网络的神经网络的神经网络神经网络的神经网络的神经网络摆着摆着摆着摆着摆着摆着摆着静止的姿势 静止的姿势姿势 姿势 姿势姿势纠正 姿势纠正在上方的上方.

更多相关视频

Author Spotlight: Automated Deep Brain Stimulation for Parkinson's Disease - Exploring the Possibilities and Challenges of Home Monitoring
06:32

Author Spotlight: Automated Deep Brain Stimulation for Parkinson's Disease - Exploring the Possibilities and Challenges of Home Monitoring

Published on: July 14, 2023

1.2K
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

1.5K

相关实验视频

Last Updated: Jun 15, 2025

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
10:23

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans

Published on: September 8, 2023

2.7K
Author Spotlight: Automated Deep Brain Stimulation for Parkinson's Disease - Exploring the Possibilities and Challenges of Home Monitoring
06:32

Author Spotlight: Automated Deep Brain Stimulation for Parkinson's Disease - Exploring the Possibilities and Challenges of Home Monitoring

Published on: July 14, 2023

1.2K
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

1.5K

科学领域:

  • 生物医学工程 生物医学工程
  • 计算机视觉 计算机视觉
  • 人与计算机的交互

背景情况:

  • 长期不适当的姿势,特别是前进的头部姿势 (FHP),与头痛,疲劳和呼吸问题有关.
  • 久坐不动的生活方式加剧FHP由于延长静态姿势维护.
  • 目前的FHP诊断依赖于不切实际的临床方法,需要实时,可访问的评估工具.

研究的目的:

  • 开发一种可访问和高效的系统,用于实时检测前置头部姿势 (FHP).
  • 为了实现持续的姿势评估,并为公共卫生效益提供纠正反.
  • 克服现有的姿势估计模型在测量FHP诊断的椎间盘角的局限性.

主要方法:

  • 使用Detectron2D和VideoPose3D算法对2D和3D人类关键点的顺序估计.
  • 利用图形卷积网络 (GCN) 在3D中分析上半身关键点的空间配置.
  • 训练GCN从3D解剖关键点数据中隐式学习FHP指标.

主要成果:

  • 通过使用身体上部关键点,GCN模型实现了78.27%的测试准确性.
  • 与基线前神经网络 (75.88%) 相比,GCN表现出优异的平衡性能 (F1得分宏:77.54%) .
  • 在各种姿势中,GCN表现出更好的概括性和平衡的FHP检测精度/回忆.

结论:

  • 基于GCN的网络可以通过3D姿势估计来学习2D图像中的FHP相关特征,用于姿势校正系统.
  • 开发的系统显示了在姿势监测和纠正方面实际应用的潜力.
  • 未来的工作将解决系统的局限性,并探索FHP检测的进一步进展.