Skip to content

Data is a huge factor in deep learning algorithms. The larger our data size, the better our model can generalize and learn. However, data preparation is a very laborious and time-consuming process. That's why I wanted to develop an application that I thought would make this stage easier. By using image processing techniques, it can track an obje…

License

Notifications You must be signed in to change notification settings

MehmetOKUYAR/Auto_video_label

Repository files navigation

Auto Video Labeling

Data is a huge factor in deep learning algorithms. The larger our data size, the better our model can generalize and learn. However, data preparation is a very laborious and time-consuming process. That's why I wanted to develop an application that I thought would make this stage easier. By using image processing techniques, it can track an object of your choice to a certain extent and saves the image and .txt file to the folder during tracking. Currently, it only works for one class and you can only label one object.

Note: it does not work very stable in videos with multivariate background

You can download the application using this link

Download - auto_video_labeling.exe

1. Open the application

Alt text

You can also run it like this python main.py

For ubuntu

requirements --> opencv-contrib-python==4.2.0.34

You should run it like this python3 auto_label.py --video video.mp4 --className label --perFrame 1 --classId 0

2. Fill in the relevant fields. If you leave it blank, the default values will be accepted.

Alt text

Load video path : Specify the video path you want to label. If you leave it blank, your camera will open.

Save per frame: Saved to one file per frame based on the value entered

Class Name : Specifies the folder name and label name to be saved. Note: It cannot contain Turkish characters.

Class Id : label id must be entered

Hotkeys

+--------------------+----------------------+
| s           | Create a rect box           |
+--------------------+----------------------+
| c           | Cancel selection rect box   |
+--------------------+----------------------+
| z           | cancel tracking operation   |
+--------------------+----------------------+
| q           | close the video             |
+--------------------+----------------------+

You can see how the program works in the gif below.

into gif

Outputs

  • A folder is created according to the class name you entered.

Alt text

  • Labels and images are saved according to the beginning of the name you enter

Alt text

  • Labels are saved in Yolo format.

Alt text

About

Data is a huge factor in deep learning algorithms. The larger our data size, the better our model can generalize and learn. However, data preparation is a very laborious and time-consuming process. That's why I wanted to develop an application that I thought would make this stage easier. By using image processing techniques, it can track an obje…

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages