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

LDIE-FDNet: Lightweight dynamic image enhancement-enabled real-time fatigue driving detection network.

PloS one·2026
Same author

Biophysical factors and management practices are key to shaping forest resilience.

Nature communications·2026
Same author

Prevalence and Correlates of Distress Detected by the Distress Thermometer and Problem List in Lung Cancer Patients: A Cross-Sectional Study.

Thoracic cancer·2025
Same author

Crop leaf disease detection with additive gated convolution and hierarchical attention fusion.

Scientific reports·2025
Same author

Multi path attention and scale aware fusion for accurate object detection in remote sensing imagery.

Scientific reports·2025
Same author

Disruptions of white matter in adolescents with internet game disorder: evidence from seed-based d mapping.

Brain imaging and behavior·2025
Same journal

DARUMA: a gateway to fast and easy prediction of intrinsically disordered regions.

PeerJ. Computer science·2026
Same journal

Alzheimer's disease detection using a quantum deep neural network with Haralick feature extraction and simulated annealing optimization.

PeerJ. Computer science·2026
Same journal

Network anomaly detection using Deep Autoencoder and parallel Artificial Bee Colony algorithm-trained neural network.

PeerJ. Computer science·2026
Same journal

An anomaly detection model for multivariate time series with anomaly perception.

PeerJ. Computer science·2026
Same journal

Retraction: A wormhole attack detection method for tactical wireless sensor networks.

PeerJ. Computer science·2026
Same journal

Evaluation of mental disorder with prioritization of its type by utilizing the bipolar complex fuzzy decision-making approach based on Schweizer-Sklar prioritized aggregation operators.

PeerJ. Computer science·2026
查看所有相关文章

相关实验视频

Updated: Sep 18, 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

4.1K

DMSA-Net:一个可变形的多尺度适应性课堂行为识别网络.

Chunyu Dong1, Jing Liu1, Shenglong Xie2

  • 1School of Computing, Xijing University, Xi'an, Shaanxi, China.

PeerJ. Computer science
|June 26, 2025
PubMed
概括
此摘要是机器生成的。

这项研究引入了一个新的网络来识别学生的课堂行为,提高远程和封闭学生的准确性. 该方法增强了特征提取和检测,在基准数据集上表现优于现有的算法.

关键词:
注意力机制注意力机制课堂行为识别识别课堂行为识别功能融合的特点是:这是一个很好的例子.对象检测检测对象检测对象检测

更多相关视频

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K
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

650

相关实验视频

Last Updated: Sep 18, 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

4.1K
A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis
05:41

A Step-by-Step Implementation of DeepBehavior, Deep Learning Toolbox for Automated Behavior Analysis

Published on: February 6, 2020

9.5K
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

650

科学领域:

  • 计算机视觉 计算机视觉
  • 人工智能的人工智能
  • 教育技术的教育技术

背景情况:

  • 准确的学生行为识别对于智能教育转型至关重要.
  • 现有的视觉算法在课堂环境中与微妙的行为,遮蔽和尺度差异作斗争.

研究的目的:

  • 提出一种可变形的多尺度自适应网络,用于增强课堂行为识别.
  • 为应对广角成像,遮蔽和尺度变化所带来的挑战.

主要方法:

  • 引入了一个可变形的自我注意 (dattention) 模块,以动态调整受感场.
  • 开发了一个多层次的注意力特征金字塔结构 (MSAFPS) 用于多层次的特征聚合.
  • 利用了智能交叉路口超过联盟 (Wise-IoU) 损失,以改进检测.

主要成果:

  • 与现有方法相比,拟议的网络表现出优越的性能.
  • 在SCB-Dataset3-S和DataMountainSCB数据集上实现了高精度.
  • 在行为识别中有效处理闭塞和尺度差异.

结论:

  • 可变形的多尺度自适应网络显著提高了课堂行为识别准确度.
  • 数据模块和MSAFPS有效地模拟分钟行为并处理尺度变化.
  • 这种方法为智能教育系统提供了强大的解决方案.