Trains a given OpenCV recognition object by preparing the given training set and call the train function of the recognizer.

The training set directories have to have the format LABEL.IGNORED_INFORMATION where the label has to be before the dot . in the directory name.

A suitable folder structure might be the following:

.
.
├── training_set
|   ├── label.1
|   |   ├── image.1.jpg
|   |   ├── . . .
|   |   └── image.n.jpg
|   ├── label.2
|   ├──  . . .
|   └── label.n
.
.

1. Open a terminal window, navigate to your project and add this python as submodule as following:

git submodule add https://github.com/n1g1r1/cv-module-face-trainer modules/face_trainer

2. Import it as python module:

from modules.face_trainer import trainer

3. Call the train function:

train(recognizer, training_set_path)

• recognizer: The given OpenCV recognizer.
• training_set_path: The path to the training set.

Or if you have to capture some webcam images first:

capture(detector, label, classifier = "lbp", resize = False, resize_factor = 0.5, make_training_set = True, training_set_size = 20, make_validation_set = False, validation_set_size = 20, training_set_path = 'data/training', validation_set_path = 'data/validation')

• detector: The detector module to use it for the face detection.
• label: The label that has to be used to make the set directory and save the images with the correct naming.
• classifier: The face classifier method that will be used for the face detection. Default: lbp.
• resize: Should the image get resized? Default: False.
• resize_factor: Resize factor, if resized. Default: 0.5.
• make_training_set: Should the algorithm make the training set? Default: True.
• training_set_size: How many images will be shot and saved. Default: 20.
• make_validation_set: Should the algorithm make the validation set? Default: False.
• validation_set_size: How many images will be shot and saved. Default: 20
• training_set_path: Where the training set gets stored. Default: data/training.
• validation_set_path: Where the validation set gets stored. Default: data/validation.