OpenCV
3.4.16
Open Source Computer Vision
|
In this tutorial you will learn how to
Set up the input video
In this tutorial, you can choose between video or list of images for the program input. As written in the help, you should specify the input video as parameter of the program. If you want to use image list as input, the image list should have formatted numbering as shown in help. In the help, it means that the image files are numbered with 4 digits (e.g. the file naming will be 0001.jpg, 0002.jpg, and so on).
You can find video samples in opencv_extra/testdata/cv/tracking https://github.com/opencv/opencv_extra/tree/master/testdata/cv/tracking
Declares the required variables
You need roi to record the bounding box of the tracked object. The value stored in this variable will be updated using the tracker object.
The frame variable is used to hold the image data from each frame of the input video or images list.
Creating a tracker object
There are at least 7 types of tracker algorithms that can be used:
Each tracker algorithm has their own advantages and disadvantages, please refer the documentation of cv::Tracker for more detailed information.
Select the tracked object
Using this function, you can select the bounding box of the tracked object using a GUI. With default parameters, the selection is started from the center of the box and a middle cross will be shown.
Initializing the tracker object
Any tracker algorithm should be initialized with the provided image data, and an initial bounding box of the tracked object. Make sure that the bounding box is valid (size more than zero) to avoid failure of the initialization process.
Update
This update function will perform the tracking process and pass the result to the roi variable.