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

HyperDC: A Non-Uniform Hypergraph Framework for Dual- and Higher-Order Drug Combination Recommendation Across Diverse Complex Diseases.

Journal of chemical information and modeling·2026
Same author

FERPIR promotes cardiomyocyte survival and attenuates cardiac remodeling after myocardial infarction.

Cell death & disease·2026
Same author

Flexible regulation of CRISPR/Cas12a activity by spatial confinement effect.

Nucleic acids research·2026
Same author

Gypenosides attenuate H<sub>2</sub>O<sub>2</sub>-induced oxidative stress in orbital fibroblasts from thyroid-associated ophthalmopathy by activating the Nrf2/ERK/HO-1 signaling pathway.

American journal of translational research·2026
Same author

Cloning and functional analysis of vitellogenin receptor gene PastVgR1 regulates reproduction in Pardosa astrigera (Araneae: Lycosidae): a potential target for mass rearing of predatory spiders in pest management.

Journal of economic entomology·2026
Same author

Inflammatory adenoma-like hepatocellular neoplasm in cirrhotic livers.

Histopathology·2026

相关实验视频

Updated: Jun 9, 2025

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K

动作识别使用基于注意力的时空VLAD网络和自适应视频序列优化.

Zhengkui Weng1,2,3, Xinmin Li4, Shoujian Xiong5

  • 1School of Automation, Qingdao University, Qingdao, 266071, China. zkweng19@163.com.

Scientific reports
|November 1, 2024
PubMed
概括

这项研究引入了基于注意力的时空VLAD网络 (AST-VLAD) 和自适应视频序列优化 (AVSO),通过更好地建模视频中的远程时间信息来改善人类行动识别.

更多相关视频

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

473
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

2.6K

相关实验视频

Last Updated: Jun 9, 2025

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

8.9K
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

473
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

2.6K

科学领域:

  • 计算机视觉 计算机视觉
  • 人工智能的人工智能
  • 机器学习 机器学习

背景情况:

  • 人类行动识别在有效捕捉视频级空间时间特征方面面临挑战.
  • 卷积神经网络 (CNN) 难以建模对复杂行动至关重要的远程时间信息.

研究的目的:

  • 开发一种基于注意力的新型时空VLAD网络 (AST-VLAD),以改进人类行为识别.
  • 为自适应视频序列优化 (AVSO) 提出一种自动方法,以增强与动作相关的表示.

主要方法:

  • AST-VLAD网络使用自我注意模型汇总了信息深度功能.
  • 适应性视频序列优化 (AVSO) 使用镜头细分和动态加权采样来优先考虑行动.
  • 自我注意力模拟了内在的时空关系,超越了简单的聚合方法.

主要成果:

  • 拟议的AST-VLAD与AVSO在公开基准上表现优越或可比.
  • 在HMDB51上达到73.1%的分类准确度,在UCF101数据集上达到96.0%.
  • 该方法有效地建模了长距离的时间依赖性,并优化了视频序列的识别.

结论:

  • 与AVSO相结合的AST-VLAD网络为人类行动识别提供了一个强大的解决方案.
  • 这种方法有效地解决了CNN在建模远程时间动态方面的局限性.
  • 拟议的方法提高了基于视频的行动分类的准确性和效率.