Skip to content

Latest commit

 

History

History
37 lines (25 loc) · 1012 Bytes

README.md

File metadata and controls

37 lines (25 loc) · 1012 Bytes

colabeler_yolo_tools

Download Colabeler(精灵标注助手) please visit http://www.jinglingbiaozhu.com/

Prerequisite:

pip install opencv-python

Make sure all pictures are jpg, use jpg_conversion.py if needed.

To change object type...

Edit index in both colabeler2yolo.py and yolo2colabeler.py

# colabeler2yolo.py
index = {'Object0':0,'Object1':1,'Object2':2,'Object3':3,'Object4':4,'Object5':5,'Object6':6,'Object7':7}

# yolo2colabeler.py
index = {0:"object0",1:"object1",2:"object2",3:"object3",4:"object4",5:"object5",6:"object6",7:"object7"}

picture folder, xml folder, yolo txt destination folder(optional)

python colabeler2yolo.py ./pic ./outputs ./outputs_yolo

picture folder, yolo txt folder, xml destination folder(optional)

python yolo2colabeler.py ./pic ./outputs_yolo ./outputs_xml