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

Demonstration of semi-automatic labeling function

Function demonstration of converting Yolo format to VOC format with one click

Attention

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

Please put classes.txt under the marked dataset folder in advance

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.

Note: This project only supports Version 5 of YOLOv5 for the time being.

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.Modify the contents of the /data/predefined_classes.txt file in the directory to your own category

5.Launching applications

python labelGo.py

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

7. 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

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