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

Extracellular Vesicles From Young Brain-Mediated Delivery of PRDX3 Ameliorates Alveolar Bone Loss in Periodontitis.

Journal of clinical periodontology·2026
Same author

Amino acid balancing and microalgal proteins (Spirulina) improve production performance and milk quality in dairy cows under low-protein diets.

NPJ science of food·2026
Same author

Flexible Prescribed-Time Optimal Control With Adaptive State-Input Constraint Bounds via Actor-Critic Learning.

IEEE transactions on neural networks and learning systems·2026
Same author

Microwave-assisted guar gum conjugation improvement in stability of pea protein high internal phase Pickering emulsion and its application in zeaxanthin delivery.

Food chemistry·2026
Same author

A Lightweight Plant Disease Detection Model for Long-Tailed Agricultural Scenarios.

Plants (Basel, Switzerland)·2026
Same author

Toward Comprehensive Information-Theoretic Multi-View Learning.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026

相关实验视频

Updated: May 24, 2025

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution
08:48

Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

Published on: September 5, 2012

11.8K

自主监督的高阶信息瓶学习尖端神经网络,以实现基于事件的强大光流估计.

Shuangming Yang, Bernabe Linares-Barranco, Yuzhu Wu

    IEEE transactions on pattern analysis and machine intelligence
    |March 3, 2025
    PubMed
    概括

    本研究介绍了SeLHIB,这是一种用于事件摄像头的新型自我监督学习算法,可以在噪音条件下增强光流估计. SeLHIB提高了稳定性和能源效率,优于现有的方法.

    科学领域:

    • 计算机视觉 计算机视觉
    • 神经形态工程的神经形态工程
    • 机器学习 机器学习

    背景情况:

    • 事件摄像头擅长高速,高动态范围的视觉感知.
    • 基于事件的光流估计的深度学习需要更好的时间特征捕获.
    • 尖端神经网络 (SNN) 提供了有效的顺序数据处理的潜力,但在概括和稳定性方面存在困难.

    研究的目的:

    • 引入SeLHIB,一种自我监督的学习算法,用于基于事件的强大的光流估计.
    • 为了利用SNN中的信息瓶理论,改进时空特征提取.
    • 在杂的视觉场景中增强概括性和强度.

    主要方法:

    • 开发了一种基于尖峰的自主监督学习算法,SeLHIB,利用信息瓶理论.
    • 员工非线性和高阶相互信息,以加强信息提取和减少冗余.
    • 在各种噪音条件下使用基于事件的摄像头输入来训练和评估算法.

    主要成果:

    • SeLHIB在不同噪音水平的光流估计中显示出显著增强的概括性和稳定性.
    • 实现了大量的节能:与模拟神经网络 (ANN) 相比减少90.44%,与尖端神经网络 (SNN) 相比减少45.70%.
    • 性能优于具有可比尺寸和架构的ANN实现,显示AEE (MVSEC) 低33.78%,RSAT (ECD) 低5.96%,RSAT (HQF) 低6.21%.

    更多相关视频

    Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
    11:18

    Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

    Published on: March 2, 2015

    10.2K
    Profiling Maternal Behavior Responses During Whole-Brain Imaging
    07:12

    Profiling Maternal Behavior Responses During Whole-Brain Imaging

    Published on: January 24, 2025

    599

    相关实验视频

    Last Updated: May 24, 2025

    Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution
    08:48

    Optical Recording of Suprathreshold Neural Activity with Single-cell and Single-spike Resolution

    Published on: September 5, 2012

    11.8K
    Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
    11:18

    Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks

    Published on: March 2, 2015

    10.2K
    Profiling Maternal Behavior Responses During Whole-Brain Imaging
    07:12

    Profiling Maternal Behavior Responses During Whole-Brain Imaging

    Published on: January 24, 2025

    599

    结论:

    • SeLHIB代表了SNNs在基于事件的视觉中的第一个自我监督的信息瓶学习策略.
    • 提出的方法有效地解决了当前SNNs在光流任务的概括性和稳定性的局限性.
    • 对于使用事件摄像头的节能和强大的视觉感知系统,SeLHIB提供了一个有前途的方向.