Bounding box labeler tool to generate the training data in the format YOLO v2 requires.
The idea is to use OpenCV so that later it uses SIFT and Tracking algorithms to make labeling easier.
I wanted this tool to give automatic suggestions for the labels! New Features Discussion
To start using the YOLO Bounding Box Tool you need to download the latest release or clone the repo:
git clone https://github.com/Cartucho/OpenLabeling
You need to install:
- Python
- OpenCV version >= 3.0
- Installation in Windows:
pip install opencv-python
- Installation in Linux
- Installation in Windows:
Step by step:
- Insert the images in the folder images/
- Insert the class list in the file class_list.txt
- Run the code:
python run.py
- You can find the bounding box files in the folder bbox_txt/
Keyboard, press:
Key | Description |
---|---|
h | help |
q | quit |
e | edges |
a/d | previous/next image |
s/w | previous/next class |
Mouse:
- Use two separate left clicks to do each bounding box
- Use the middle mouse to zoom in and out
- Use double click to select a bounding box
- Right click to quickly delete a bounding box
Generating XML Files for Darkflow from .txt files generated by OpenLabeling
USAGE: generate_xml.py [-h] [--format {yolo,voc}] [--bbox_txt BBOX_TXT]
[--img_dir IMG_DIR] [--class_list CLASS_LIST]
[--save_dir SAVE_DIR]
Generate XML Files from .txt file
optional arguments:
-h, --help show this help message and exit
--format {yolo,voc} Bounding box format | Deafult: yolo
--bbox_txt BBOX_TXT Path to bbox_txt dir generated by OpenLabeling | Deafult: bbox_txt/
--img_dir IMG_DIR Path to Images folder | Default: images/
--class_list CLASS_LIST Path to class_list.txt file | Default: class_list.txt
--save_dir SAVE_DIR Where to save XML Files. | Default: annotations_xml/
-
João Cartucho - Please give me your feedback: [email protected]
Feel free to contribute