this opencv app for live language sign detection
- we gathered images with opencv through the webcam
- we labeled the data and augmented it with python modules
- we trained model for the job
- clone the repo
git clone repo
cd repo
pip install -r requirements.txt- structure the project
python scripts/create_dirs.py- capture the data
python scripts/capture.py-
label the data open the notebook and run the labelme cell to triger the api and you have to labels the images you select the resized_data/images dir and at the top left corner from the drop list select change output dir and select resized_data/labels and from the drop list click save automaticly after that from the next drop list select draw rectangle and start label your images
-
split the data
python scripts/split.py - augment the data
python scripts/augment.py -
train the model open the note book and run the cells to train the model
-
detect through the webcam
python scripts/live-detection.py you will be able to detect some thing like that
