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

Improving Translational Accuracy02:07

Improving Translational Accuracy

2.6K
2.6K

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

Identifying the window of aggressive postpartum breast cancer based on the 21-gene Oncotype DX® test in women with HR+, HER2-negative breast cancer.

NPJ breast cancer·2026
Same author

A multitransmitter, synaptic specification-segregated calyx-like synapse linking pontine pituitary adenylate cyclase-activating polypeptide neurons to forebrain-extended amygdala.

PNAS nexus·2026
Same author

Multiplex on-chip detection of <i>Aspergillus</i> by integrated ultrasonication-based bead beating lysis and magnetic beads direct amplification.

Frontiers in bioengineering and biotechnology·2026
Same author

Understanding Barriers to Colon Cancer Screening Among Individuals Experiencing Housing Insecurity in Los Angeles.

Cureus·2026
Same author

Association between the telomerase rs2736100_CC genotype and a lower risk of chronic hepatitis C in Chinese males.

Molecular biology reports·2025
Same author

Ventral Tegmental Area Dopamine Neurons Regulated by the Parabrachial Nucleus Mediate Long-Term Aversive Memory.

Biological psychiatry·2025

相关实验视频

Updated: Jun 29, 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

9.0K

强大的细粒度视觉识别与邻居注意力标签校正

Shunan Mao, Shiliang Zhang

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
    |March 28, 2024
    PubMed
    概括

    本研究引入了一个邻居注意力标签校正 (NALC) 模型,用于在深度学习中修复噪音标签,以实现细粒度的视觉识别. NALC显著提高了标签准确性,并改善了模型在各种识别任务上的性能.

    科学领域:

    • 计算机科学 计算机科学
    • 人工智能的人工智能
    • 机器学习 机器学习

    背景情况:

    • 为了精细的视觉识别,深度学习需要广泛,准确的注释.
    • 现实世界的数据收集通常会引入标签噪声,阻碍模型的性能.
    • 现有的方法在细粒度数据集中与固有的噪音作斗争.

    研究的目的:

    • 在深度模型培训中解决标签噪声的挑战,以实现细粒度视觉识别.
    • 在培训阶段提出一种用于纠正噪音标签的新方法.
    • 在有标签噪音的情况下,提高深度学习模型的稳定性和准确性.

    主要方法:

    • 提出了邻居注意力标签纠正 (NALC) 模型用于自动标签纠正.
    • 使用一个具有验证批次的元学习框架来纠正训练批次标签.
    • 引入了一个嵌套优化算法来提高元学习效率.
    • 实施了NALC以完善培训批次内的标签准确性.

    主要成果:

    • 从70%提高到98%以上,显著提高了标签准确度.
    • 在细粒度图像检索任务中,平均平均精度 (mAP) 超过现有方法高达13.4%.
    • 在杂的语义细分数据集上,实现了对欧盟平均交叉点 (mIOU) 7.8% 的改进.

    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

    530
    Targeted Labeling of Neurons in a Specific Functional Micro-domain of the Neocortex by Combining Intrinsic Signal and Two-photon Imaging
    11:24

    Targeted Labeling of Neurons in a Specific Functional Micro-domain of the Neocortex by Combining Intrinsic Signal and Two-photon Imaging

    Published on: December 12, 2012

    13.7K

    相关实验视频

    Last Updated: Jun 29, 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

    9.0K
    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

    530
    Targeted Labeling of Neurons in a Specific Functional Micro-domain of the Neocortex by Combining Intrinsic Signal and Two-photon Imaging
    11:24

    Targeted Labeling of Neurons in a Specific Functional Micro-domain of the Neocortex by Combining Intrinsic Signal and Two-photon Imaging

    Published on: December 12, 2012

    13.7K
  • 证明了对各种模拟和实际噪音类型的强度.

    • 结论:

    • NALC有效地纠正噪音标签,增强学习的图像表示.
    • 拟议的方法在各种细粒度视觉识别任务中提供了实质性的性能提升.
    • NALC提供了一个强大的解决方案,用于训练有噪音数据集的深度学习模型.