Search research articles

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 Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

Parallel Processing

Parallel Processing

The brain processes sensory information rapidly due to parallel processing, which involves sending data across multiple neural pathways at the same time. This method allows the brain to manage various sensory qualities, such as shapes, colors, movements, and locations, all concurrently. For instance, when observing a forest landscape, the brain simultaneously processes the movement of leaves, the shapes of trees, the depth between them, and the various shades of green. This enables a quick and...

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

Effect of control frequency on closed-loop electrical muscle stimulation for biceps-triceps in healthy participants: a pilot study.

Scientific reports·2026

Same author

Multidetector computed tomography angiography predicts the need for bronchial artery embolization in hemoptysis: A retrospective cohort study of 32 patients.

Respiratory medicine case reports·2025

Same author

Ultra-low-latency 3D measurement with a 1D approximate de Bruijn pattern.

Optics letters·2025

Same author

Klinefelter syndrome diagnosed at autopsy and small-cell lung carcinoma.

Respiratory medicine case reports·2025

Same author

High-Speed Tracking with Mutual Assistance of Feature Filters and Detectors.

Sensors (Basel, Switzerland)·2023

Same author

Edge-Supervised Linear Object Skeletonization for High-Speed Camera.

Sensors (Basel, Switzerland)·2023

Search research articles

相关实验视频

Updated: Jan 17, 2026

FIM Imaging and FIMtrack: Two New Tools Allowing High-throughput and Cost Effective Locomotion Analysis

FIM Imaging and FIMtrack: Two New Tools Allowing High-throughput and Cost Effective Locomotion Analysis

Published on: December 24, 2014

基于智能和实时图像处理的融合的高速多重对象跟踪.

Yuki Kawawaki¹, Yuji Yamakawa²

¹Graduate School of Engineering, The University of Tokyo, Tokyo 153-8505, Japan.

Sensors (Basel, Switzerland)

|September 19, 2025

概括

此摘要是机器生成的。

这项研究引入了一种新的高速多重物体跟踪 (MOT) 系统,可以平衡速度和准确性. 混合方法显著提高计算机视觉应用程序的实时性能.

关键词:

深度学习是一种深度学习.高速度的处理速度.混合追踪追踪是混合的追踪.多重处理多重处理多个对象跟踪多个对象跟踪追踪器管理管理

更多相关视频

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques

Published on: March 12, 2019

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

相关实验视频

Last Updated: Jan 17, 2026

FIM Imaging and FIMtrack: Two New Tools Allowing High-throughput and Cost Effective Locomotion Analysis

FIM Imaging and FIMtrack: Two New Tools Allowing High-throughput and Cost Effective Locomotion Analysis

Published on: December 24, 2014

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques

Published on: March 12, 2019

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

科学领域:

计算机视觉计算机视觉
机器人技术机器人技术机器人技术
人工智能的人工智能

背景情况:

多重对象跟踪 (MOT) 对于自动驾驶和监控等应用至关重要.
现有的MOT方法通常优先考虑关联而不是检测速度,从而限制实时性能.
需要MOT系统来平衡速度,准确性和稳定性.

研究的目的:

开发一套高速的MOT系统,在不牺牲跟踪精度的情况下提高实时性能.
调查加速检测对MOT系统整体效率的影响.
提出一种新的混合追踪框架和追踪器管理策略.

主要方法:

一个混合跟踪框架,将低频深度学习检测与经典的高速跟踪相结合.
一个基于检测标签的策略,用于管理对象跟踪.
在六个场景中使用高速摄像机数据进行评估,并与七种最先进的 (SOTA) 方法进行比较.

主要成果:

实现了高速率:高达470fps (2个对象),243fps (3个对象) 和178fps (4个对象).
通过高精度检测,在MOTA,IDF1和HOTA中获得最高分数.
证明有效的长期协会用于高速跟踪,即使检测准确度较低.

结论:

拟议的系统为高速MOT提供了一个实用和高效的基线.
多处理架构推动了MOT研究,特别是对于具有异步模块的系统.
混合方法有效平衡实时性能,跟踪精度和稳定性.