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

A comprehensive examination and analysis of the effectiveness and safety of finerenone for the treatment of diabetic kidney disease: a systematic review and meta-analysis.

Frontiers in endocrinology·2025
Same author

Genome-wide identification and expression patterns of uridine diphosphate (UDP)-glycosyltransferase genes in the brown planthopper, Nilaparvata lugens.

Comparative biochemistry and physiology. Part D, Genomics & proteomics·2025
Same author

Myeloid-Derived Suppressor Cell Accumulation Drives Intestinal Fibrosis through mCCL6/hCCL15 Chemokine-Mediated Fibroblast Activation.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2024
Same author

Enhanced Thermal Safety of Hydrophobic SiO<sub>2</sub> Aerogels Through Introduction of Layered Double Oxides.

Gels (Basel, Switzerland)·2024
Same author

Detection by super-resolution microscopy of viral proteins inside bloodborne extracellular vesicles.

Extracellular vesicles and circulating nucleic acids·2024
Same author

Bullying victimization and self-harm in adolescents: The roles of emotion regulation and bullying peer norms.

Child abuse & neglect·2024

相关实验视频

Updated: Jul 23, 2025

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

7.7K

基于卷积结构和基于滑窗的注意力机制的凝视估计.

Yujie Li1,2, Jiahui Chen1, Jiaxin Ma1

  • 1School of Artificial Intelligence, Guilin University of Electronic Technology, Guilin 541004, China.

Sensors (Basel, Switzerland)
|July 14, 2023
PubMed
概括

这项研究引入了Swin变压器模型,以改进视线估计. 混合Res-Swin-GE模型显著优于现有方法,增强对人类注意力和认知状态的理解.

关键词:
卷积神经网络 (CNN) 是一种神经网络.深度学习是一种深度学习.凝视的估计估计.自己注意力机制机制.斯温变压器 的变压器

更多相关视频

Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior
07:09

Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior

Published on: November 14, 2018

10.7K
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

3.8K

相关实验视频

Last Updated: Jul 23, 2025

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

7.7K
Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior
07:09

Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior

Published on: November 14, 2018

10.7K
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

3.8K

科学领域:

  • 计算机视觉 计算机视觉
  • 人工智能的人工智能
  • 人与计算机的交互

背景情况:

  • 人的视线方向对于理解注意力和认知状态至关重要.
  • 卷积神经网络 (CNN) 在全球建模中显示了目光估计的局限性.
  • 变压器模型提供了进步,但可以降低当地的空间信息.

研究的目的:

  • 引入Swin变压器架构,以增强视线估计.
  • 解决现有的CNN和变压器模型在捕捉多尺度特征和局部空间细节方面的局限性.

主要方法:

  • 开发了两个基于Swin变压器的目光估计模型:SwinT-GE (纯Swin变压器) 和Res-Swin-GE (混合CNN-Swin变压器).
  • Res-Swin-GE集成了卷积结构,以取代SwinT-GE中的切片和映射机制.
  • 在MpiiFaceGaze和Eyediap数据集上评估模型.

主要成果:

  • Res-Swin-GE显著超过了纯SwinT-GE模型的表现.
  • 混合Res-Swin-GE模型在MpiiFaceGaze数据集上表现出强大的竞争力.
  • 在 Eyediap 数据集上,与最先进的方法相比,实现了 7.5% 的性能改进.

结论:

  • 斯温变压器架构,特别是混合Res-Swin-GE,为推进凝视估计提供了一个有希望的方向.
  • 混合方法有效地平衡全球和本地特征学习,以获得卓越的性能.
  • 这项工作有助于通过凝视跟踪更准确地分析人类行为和认知状态.