Skip to content

sangyy/trackingROI

Repository files navigation

Project Title

GitHub license

Learn object tracking. We will use our mouse to select an object and track it using different methods that opencv has to offer.

Demo-Preview

Object Tracking Basic Code Initializing the Tracker Tracking the Object Drawing the Bounding Box Types of Trackers Video Tutorial Complete Code

Random GIF

Table of contents

Installation

(Back to top)

Here is a sample instruction:

To use this project, first clone the repo on your device using the command below:

git init

git clone https://github.com/sangyy/trackingROI

Usage

(Back to top)

八种Tracker包括:

BOOSTING Tracker:和Haar cascades(AdaBoost)背后所用的机器学习算法相同,但是距其诞生已有十多年了。这一追踪器速度较慢,并且表现不好,但是作为元老还是有必要提及的。(最低支持OpenCV 3.0.0)

MIL Tracker:比上一个追踪器更精确,但是失败率比较高。(最低支持OpenCV 3.0.0)

KCF Tracker:比BOOSTING和MIL都快,但是在有遮挡的情况下表现不佳。(最低支持OpenCV 3.1.0)

CSRT Tracker:比KCF稍精确,但速度不如后者。(最低支持OpenCV 3.4.2)

MedianFlow Tracker:在报错方面表现得很好,但是对于快速跳动或快速移动的物体,模型会失效。(最低支持OpenCV 3.0.0)

TLD Tracker:我不确定是不是OpenCV和TLD有什么不兼容的问题,但是TLD的误报非常多,所以不推荐。(最低支持OpenCV 3.0.0)

MOSSE Tracker:速度真心快,但是不如CSRT和KCF的准确率那么高,如果追求速度选它准没错。(最低支持OpenCV 3.4.1)

GOTURN Tracker:这是OpenCV中唯一一深度学习为基础的目标检测器。它需要额外的模型才能运行,本文不详细讲解。(最低支持OpenCV 3.2.0)

我个人的建议:

如果追求高准确度,又能忍受慢一些的速度,那么就用CSRT

如果对准确度的要求不苛刻,想追求速度,那么就选KCF

纯粹想节省时间就用MOSSE

License

(Back to top)

GitHub license

Footer

(Back to top)

Leave a star in GitHub and share this guide if you found this helpful.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published