labelGo

Guide Language:简体中文

A graphical Semi-automatic annotation tool based on labelImg and YOLOv5

Semi-automatic annotation of datasets by existing yolov5 pytorch models

News

labelGo now supports the latest version of YOLOv5, and automatic classes.txt file generation

Demonstration of semi-automatic labeling function

Demonstration of converting Yolo format to VOC format with one click

Note

If there is a problem, please put it forward in the issue.

The annotation file is saved in the same location as the picture folder.

Recommended version of python: python 3.8.

Recommended for conda environments.

The item is completely free and it is forbidden to sell the item in any way.

This project has support for the latest version of YOLOv5, if you need to use an older version that supports YOLOv5 version5, you can find the source code in Release.

Installation and use

1.Fetching projects from git

git clone https://github.com/cnyvfang/labelGo-Yolov5AutoLabelImg.git

2.Switching the operating directory to the project directory

cd labelGo-Yolov5AutoLabelImg

3.Installation environment

pip install -r requirements.txt

4.Launching applications

python labelGo.py

5. Click on the "Open directory" button to select the folder where the images are stored

6. Click on the "Auto Annotate" button to confirm that the information is correct and then select the trained yolov5 pytorch model to complete the auto annotation

7. Adjust the automatic annotation results according to the actual requirements and save them

Acknowledgements

Thanks to tangtang666 for submitting support for the latest version of YOLOv5

Thanks to Iceprism for fixing the bugs in the Chinese version.