/Kitchen-YOLO-Dataset

Primary LanguagePythonOtherNOASSERTION

Kitchen YOLO Datset

This datset contains the Label Studio Rectangular labels for the following object classes inside respective container types:

Translucent Containers:

  • Bell Pepper
  • Pasta
  • Chilli
  • Garlic
  • Cheese
  • Butter

Opaque container:

  • Mushrooms
  • Peas and Carrots
  • Green Beans
  • Tomato Sauce
  • Corn

To train the dataset using your favorite deep neural Network/Frameworks, you can use Label Studio and export the rectangular bounding boxes in a wide variety of formats.

Dataset Statistics

Class Image Count by Class
Cheese 18
Peas and Carrots 25
Butter 32
Green Beans 34
Corn 48
Mushrooms 52
Garlic 56
Bell Pepper 57
Tomato Sauce 58
Chilli 60
Pasta 62

Install and Launch Label Studio

To install label studio to work with this dataset, install label-studio in an Poetry virtual environment:

poetry install

Then subsequently activate the virtual environment.

cd DIRECTORY_OF_THIS_REPOSITORY
source .venv/bin/activate
export LABEL_STUDIO_LOCAL_FILES_SERVING_ENABLED=true
export LABEL_STUDIO_LOCAL_FILES_DOCUMENT_ROOT=$PWD
export DATA_UPLOAD_MAX_NUMBER_FILES=10000000 
export LABEL_STUDIO_DISABLE_SIGNUP_WITHOUT_LINK=true
echo $LABEL_STUDIO_LOCAL_FILES_SERVING_ENABLED
echo $LABEL_STUDIO_LOCAL_FILES_DOCUMENT_ROOT
label-studio start --username hmi2@gmail.com --password test123

After running the label Studio server, import the images from the my_images folder and the annotations from final_export file into the label studio server. After importing the data, you can export it to whichever format you want.