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

HK2/VDAC1 Axis Regulates Malignant Progression of Esophageal Squamous Cell Carcinoma.

Cancer medicine·2026
Same author

Fabrication of Indole-3-propionic acid loaded hydrogels for spinal cord injury repair.

Biomaterials advances·2026
Same author

Zn-Doped CeO<sub>2</sub> Nanorod-Supported Ni-Co Bimetallic Catalysts for Enhanced Photothermal Catalytic CH<sub>4</sub>-CO<sub>2</sub> Reforming.

The journal of physical chemistry letters·2026
Same author

Balloon-occluded hepatic arterial infusion for unresectable hepatocellular carcinoma: a phase II trial interim analysis.

Frontiers in oncology·2026
Same author

ASAP3 activates the NF-κB signaling pathway to promote the progression of esophageal squamous cell carcinoma.

Discover oncology·2026
Same author

Quantitative CT Imaging Features Associated With High Risk of Elevated Intracranial Pressure in Children Under 2 Years of Age With Craniosynostosis: A Patient-Control Study.

The Journal of craniofacial surgery·2026

相关实验视频

Updated: Jun 19, 2025

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
10:23

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans

Published on: September 8, 2023

2.7K

用于数字图书馆的面部长度和角度特征识别.

Shuangyan Li1, Min Ji2, Ming Chen2

  • 1School of Art and Design, Qingdao University of Technology, Qingdao, China.

PloS one
|July 24, 2024
PubMed
概括

本研究引入了一种使用数字图书馆的长度和角度特征的新面部识别方法. 以注意力为基础的方法在识别各种面部表情方面实现了高精度,增强了生物识别技术.

科学领域:

  • 计算机科学 计算机科学
  • 生物识别信息 生物识别信息
  • 人工智能的人工智能

背景情况:

  • 面部识别是一个成熟的生物识别技术,但准确性仍然是一个挑战.
  • 数字图书馆需要强大的面部特征识别来改善服务.

研究的目的:

  • 为数字图书馆提出一种使用长度和角度特征的新型面部识别方法.
  • 用注意力机制提高面部表情识别的准确性和稳定性.

主要方法:

  • 开发了一个面部动作网络架构,结合了注意力机制.
  • 探索了一个网络架构,重点关注面部表情的长度和角度特征.
  • 构建了一个端到端的框架,利用注意面部特征点.

主要成果:

  • 在FER-2013数据集上,七个常见表达式的平均识别率为97.28%99.97%.
  • 报告了幸福和惊喜 (99.97%) 的最高识别率和愤怒,恐惧和中立 (97.18%) 的较低识别率.
  • 在面部表情识别方面表现出高精度和稳定性,特别是在复杂的环境中.

结论:

  • 提出的基于注意力的方法显著提高了面部表情识别的准确性和稳定性.

更多相关视频

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

820
Quantification of Orofacial Phenotypes in Xenopus
09:26

Quantification of Orofacial Phenotypes in Xenopus

Published on: November 6, 2014

9.7K

相关实验视频

Last Updated: Jun 19, 2025

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans
10:23

Author Spotlight: Three-Dimensional Cephalometric Landmark Annotation Demonstration on Human Cone Beam Computed Tomography Scans

Published on: September 8, 2023

2.7K
Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

820
Quantification of Orofacial Phenotypes in Xenopus
09:26

Quantification of Orofacial Phenotypes in Xenopus

Published on: November 6, 2014

9.7K
  • 该方法为数字图书馆和其他应用程序提供可靠的技术支持.
  • 这项研究促进了数字图书馆服务和用户体验的面部识别技术的进步.