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

Predicting Emergency Department Patient Arrivals at Hospitals Using Machine Learning Techniques.

Healthcare (Basel, Switzerland)·2026
Same author

An Enhanced Team-Oriented Swarm Optimization Algorithm (ETOSO) for Robust and Efficient High-Dimensional Search.

Biomimetics (Basel, Switzerland)·2025
Same author

A comparative study of neural network architectures for vital signs monitoring based on the national early warning systems (NEWS).

Health informatics journal·2025
Same author

Compact Wideband Four-Port MIMO Antenna for Sub-6 GHz and Internet of Things Applications.

Micromachines·2022
Same author

Metamaterial-Inspired Electrically Compact Triangular Antennas Loaded with CSRR and 3 × 3 Cross-Slots for 5G Indoor Distributed Antenna Systems.

Micromachines·2022
Same author

ABET Accreditation During and After COVID19 - Navigating the Digital Age.

IEEE access : practical innovations, open solutions·2022

相关实验视频

Updated: Mar 15, 2026

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

5.6K

一个智能实时系统用于连续的沙特手语的句子级识别,使用基于地标的时间建模.

Adel BenAbdennour1, Mohammed Mukhtar1, Osama Almolike1

  • 1Department of Electrical Engineering, Faculty of Engineering, Islamic University of Madinah, Madinah 42351, Saudi Arabia.

Sensors (Basel, Switzerland)
|March 14, 2026
PubMed
概括

本研究介绍了实时沙特手语 (SSL) 识别系统,该系统将连续的标志翻译成口语阿拉伯语. 该系统达到94.2%的准确性,弥合了聋人和听力障碍者之间的沟通差距.

关键词:
双向长期短期记忆 双向长期短期记忆沙特手语是沙特的手语.人工智能的人工智能是人工智能.辅助技术的使用.深度学习是一种深度学习.身体残疾就是残疾.大型语言模型在句子级别的识别.标语是指手语的使用方式.时间建模时间建模

更多相关视频

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

2.1K
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
07:14

Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models

Published on: December 23, 2025

471

相关实验视频

Last Updated: Mar 15, 2026

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

5.6K
Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

2.1K
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
07:14

Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models

Published on: December 23, 2025

471

科学领域:

  • 计算机科学 计算机科学
  • 人工智能的人工智能
  • 人与计算机的交互

背景情况:

  • 在手语使用者和听力社区之间仍然存在沟通障碍,特别是在缺乏自动翻译的地区.
  • 由于依赖沙特手语 (SSL) 和实时翻译系统的稀缺,沙特阿拉伯面临着扩大通信挑战.

研究的目的:

  • 开发和评估一个实时,端到端的系统,用于持续的沙特手语 (SSL) 句子识别.
  • 直接将认可的SSL句子映射到自然的口语阿拉伯语输出.
  • 解决沙特阿拉伯聋人和听障人士的沟通缺口.

主要方法:

  • 使用MediaPipe Holistic从视频流中提取时空地标特征.
  • 采用双向长期短期记忆 (BiLSTM) 网络,在ArabSign (ArSL) 数据集上进行训练,以进行句子级分类.
  • 整合了一个基于置的细分机制,以实现自然,不间断的签名,并使用Leave-One-Signer-Out (LOSO) 交叉验证进行评估.

主要成果:

  • 使用LOSO协议实现了94.2%的平均句子级准确度,超过了基线 (92.07%).
  • 证明了强大的泛化和实时性能,适合交互式应用程序.
  • 基于LLM的可选改进阶段被整合到阿拉伯语输出中,以提高语言流性.

结论:

  • 直接句子级建模,结合基于地标的功能和实时细分,为持续的SSL识别提供了有效的解决方案.
  • 开发的系统显著改善了SSL用户的实时通信可访问性.
  • 该方法为开发先进的手语翻译技术提供了实际基础.