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

Edge-Based Vision-Language Assistive System for the Visually Impaired: A Quantized VLM Approach.

IEEE transactions on neural systems and rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society·2026
Same author

Hierarchical human activity recognition with fusion of audio and multiple inertial sensor modalities.

Scientific reports·2025
Same author

Enhancing healthcare data privacy and interoperability with federated learning.

PeerJ. Computer science·2025
Same author

EAV: EEG-Audio-Video Dataset for Emotion Recognition in Conversational Contexts.

Scientific data·2024
Same author

A decision support system for assessing landfill performance.

Waste management (New York, N.Y.)·2009

相关实验视频

Updated: Jun 28, 2025

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

3.7K

GLULA:基于注意力的线性模型,用于通过可穿戴传感器高效地识别人类活动.

Aldiyar Bolatov1, Aigerim Yessenbayeva1, Adnan Yazici1

  • 1Department of Computer Science, Nazarbayev University, Astana, Kazakhstan.

Wearable technologies
|April 15, 2024
PubMed
概括

研究人员开发了GLULA,这是一种使用身上的传感器进行人类活动识别 (HAR) 的新框架. 这种高效的模型提高了实时应用程序的速度和内存使用率,在基准数据集上表现优于现有的方法.

科学领域:

  • 生物医学工程 生物医学工程
  • 计算机科学 计算机科学
  • 机器学习 机器学习

背景情况:

  • 身体佩戴的传感器对于监控患者康复和控制设备至关重要.
  • 精确的人类活动识别 (HAR) 需要捕获复杂的时空数据依赖.
  • 优化模型效率 (内存,推理时间) 对实时和移动应用程序至关重要.

研究的目的:

  • 为了介绍GLULA,一个新的,高效的,强大的建筑HAR.
  • 为了解决现有模型在速度和内存使用方面的局限性.
  • 为了提高HAR性能,特别是在数据有限的场景中.

主要方法:

  • 开发了GLULA,一个独特的架构,结合了封闭的卷积网络,分支卷积和线性自我注意力.
  • 利用多元组混合作为一种增强技术,以提高有限数据的性能.
  • 在五个基准数据集 (PAMAP2,SKODA,OPPORTUNITY,DAPHNET,USC-HAD) 上进行了广泛的实验.

主要成果:

  • 在5个基准数据集中,GLULA在4个基准数据集中表现优于最近的模型.
  • 拟议的架构在与之比较的最先进模型中实现了最低的参数数量.
  • GLULA的推断时间接近最先进的状态,表明高效率.
关键词:
深度学习是一种深度学习.人类活动的认可 人类活动的认可人与机器人的互动线性自我注意力线性自我注意力

更多相关视频

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K
Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
07:47

Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification

Published on: February 14, 2018

11.2K

相关实验视频

Last Updated: Jun 28, 2025

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

3.7K
Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
06:49

Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

Published on: December 11, 2015

8.9K
Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
07:47

Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification

Published on: February 14, 2018

11.2K

结论:

  • GLULA 提供了一种高效和有效的解决方案,用于 HAR 任务,使用身体佩戴的传感器数据.
  • 该架构的设计平衡了高性能与降低计算成本.
  • 在康复和其它领域,GLULA为实时HAR应用提供了一个有前途的进步.