face-mask-classifier

SVM facemask classifier

Pretrained Model used: Keras-facenet model () This model is used to detect the faces in a image and create a embedding vector of the faces which would be passed to a SVM classifier. The embedding vector is of dimension 512.

Download the SVM pretrained model https://github.com/Baskar-t/face-mask-classifier/tree/main/models

Class mask_predictor

image - path of the image file to predict the mask

pretrained path of the SVM model

def pred_mask(image,svm_model_path) ...... return img, box_list, label_list, proba_list

img - image vector

box list - list of image box of faces

label_list - list of labels for the predicted face images (Mask , No Mask)

proba_list - list of probability of predicted labels

Usage:

mask_predictor.pred_mask(image path, path_svm_model)