preprocessingEmbeddings.py

from imutils import paths
import numpy as np
import imutils
import pickle
import cv2
import os


def embeddings():

	dataset = "dataset"

	embeddingFile = "output/embeddings.pickle" #initial name for embedding file
	embeddingModel = "openface_nn4.small2.v1.t7" #initializing model for embedding Pytorch

	#initialization of caffe model for face detection
	prototxt = "model/deploy.prototxt"
	model =  "model/res10_300x300_ssd_iter_140000.caffemodel"

	#loading caffe model for face detection
	#detecting face from Image via Caffe deep learning
	detector = cv2.dnn.readNetFromCaffe(prototxt, model)

	#loading pytorch model file for extract facial embeddings
	#extracting facial embeddings via deep learning feature extraction
	embedder = cv2.dnn.readNetFromTorch(embeddingModel)

	#gettiing image paths
	imagePaths = list(paths.list_images(dataset))

	#initialization
	knownEmbeddings = []
	knownNames = []
	total = 0
	conf = 0.5

	#we start to read images one by one to apply face detection and embedding
	for (i, imagePath) in enumerate(imagePaths):
		print("Processing image {}/{}".format(i + 1,len(imagePaths)))
		name = imagePath.split(os.path.sep)[-2]
		image = cv2.imread(imagePath)
		image = imutils.resize(image, width=600)
		(h, w) = image.shape[:2]
		#converting image to blob for dnn face detection
		imageBlob = cv2.dnn.blobFromImage(
			cv2.resize(image, (300, 300)), 1.0, (300, 300),(104.0, 177.0, 123.0), swapRB=False, crop=False)

		#setting input blob image
		detector.setInput(imageBlob)
		#prediction the face
		detections = detector.forward()

		if len(detections) > 0:
			i = np.argmax(detections[0, 0, :, 2])
			confidence = detections[0, 0, i, 2]

			if confidence > conf:
				#ROI range of interest
				box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
				(startX, startY, endX, endY) = box.astype("int")
				face = image[startY:endY, startX:endX]
				(fH, fW) = face.shape[:2]

				if fW < 20 or fH < 20:
					continue
				#image to blob for face
				faceBlob = cv2.dnn.blobFromImage(face, 1.0 / 255, (96, 96), (0, 0, 0), swapRB=True, crop=False)
				#facial features embedder input image face blob
				embedder.setInput(faceBlob)
				vec = embedder.forward()
				knownNames.append(name)
				knownEmbeddings.append(vec.flatten())
				total += 1

	print("Embedding:{0} ".format(total))
	data = {"embeddings": knownEmbeddings, "names": knownNames}
	f = open(embeddingFile, "wb")
	f.write(pickle.dumps(data))
	f.close()
	print("Process Completed")

embeddings()