Created on Mon Dec 03 11:15:45 2018
@author: keyur-r
python facial_landmarks.py -i <image> -l <> -w <> -d <> -p <> -m <> -t <>
l -> hog or cnn or dl
w -> model path for facial landmarks (shape_predictor_68_face_landmarks.dat)
d -> cnn trained model path (mmod_human_face_detector.dat)
p -> Caffe prototype file for dnn module (deploy.prototxt.txt)
m -> Caffe trained model weights path (res10_300x300_ssd_iter_140000.caffemodel)
t -> Thresold to filter weak face in dnn'''
import numpy as np
import dlib
import cv2
import argparse
import os
from image_utility import save_image, generate_random_color, draw_border
from imutils import face_utils
def hog_landmarks(image, gray):
faces_hog = face_detector(gray, 1)
# HOG + SVN
for (i, face) in enumerate(faces_hog):
# Finding points for rectangle to draw on face
x, y, w, h = face.left(), face.top(), face.width(), face.height()
# Drawing simple rectangle around found faces
cv2.rectangle(image, (x, y), (x + w, y + h), generate_random_color(), 2)
# Make the prediction and transfom it to numpy array
shape = predictor(gray, face)
shape = face_utils.shape_to_np(shape)
# Draw on our image, all the finded cordinate points (x,y)
for (x, y) in shape:
cv2.circle(image, (x, y), 2, (0, 255, 0), -1)
def cnn_landmarks(image, gray):
faces_cnn = face_detector(gray, 1)
# CNN
for (i, face) in enumerate(faces_cnn):
# Finding points for rectangle to draw on face
x, y, w, h = face.rect.left(), face.rect.top(), face.rect.width(), face.rect.height()
# Drawing simple rectangle around found faces
cv2.rectangle(image, (x, y), (x + w, y + h), generate_random_color(), 2)
# Make the prediction and transfom it to numpy array
shape = predictor(gray, face.rect)
shape = face_utils.shape_to_np(shape)
# Draw on our image, all the finded cordinate points (x,y)
for (x, y) in shape:
cv2.circle(image, (x, y), 2, (0, 255, 0), -1)
def dl_landmarks(image, gray, h, w):
# # This is based on SSD deep learning pretrained model
# https://docs.opencv.org/trunk/d6/d0f/group__dnn.html#ga29f34df9376379a603acd8df581ac8d7
inputBlob = cv2.dnn.blobFromImage(cv2.resize(
image, (300, 300)), 1, (300, 300), (104, 177, 123))
face_detector.setInput(inputBlob)
detections = face_detector.forward()
for i in range(0, detections.shape[1]):
# Probability of prediction
prediction_score = detections[0, 0, i, 2]
if prediction_score < args.thresold:
continue
# compute the (x, y)-coordinates of the bounding box for the
# object
box = detections[0, 0, i, 3:7] * np.array([w, h, w, h])
(x1, y1, x2, y2) = box.astype("int")
# For better landmark detection
y1, x2 = int(y1 * 1.15), int(x2 * 1.05)
# Make the prediction and transfom it to numpy array
shape = predictor(gray, dlib.rectangle(left=x1, top=y1, right=x2, bottom=y2))
shape = face_utils.shape_to_np(shape)
cv2.rectangle(image, (x1, y1), (x2, y2), generate_random_color(), 2)
# Draw on our image, all the finded cordinate points (x,y)
for (x, y) in shape:
cv2.circle(image, (x, y), 2, (0, 255, 0), -1)
def face_detection(image):
# Converting the image to gray scale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
# The 1 in the second argument indicates that we should upsample the image
# 1 time. This will make everything bigger and allow us to detect more
# faces.
# write at the top left corner of the image
img_height, img_width = image.shape[:2]
if model == 'hog':
hog_landmarks(image, gray)
elif model == 'cnn':
cnn_landmarks(image, gray)
else:
dl_landmarks(image, gray, img_height, img_width)
cv2.putText(image, "68 Pts - {}".format(model), (img_width - 200, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.5,
generate_random_color(), 2)
save_image(image)
# Show the image
cv2.imshow("Facial Landmarks", image)
cv2.waitKey(0)
cv2.destroyAllWindows()
if __name__ == "__main__":
HOME ="C:/study folder/IP project/facial-landmarks-master/IMG2.PNG"
# handle command line arguments
ap = argparse.ArgumentParser()
ap.add_argument('-i', '--image', required=True, default="C:/study folder/IP project/facial-landmarks-master/IMG2.PNG" ,help='Path of image')
ap.add_argument("-l", "--learning", default="hog",
help="Which learning model from hog/dl/cnn to use for FaceDetection!")
ap.add_argument('-w', '--weights',
default='C:/study folder/IP project/facial-landmarks-master/shape_predictor_68_face_landmarks.dat', help='Facial Landmarks Model')
ap.add_argument('-d', '--data', help='CNN trained model',
default='./mmod_human_face_detector.dat')
ap.add_argument("-p", "--prototxt", default="C:/study folder/IP project/facial-landmarks-master/deploy.prototxt",
help="Caffe 'deploy' prototxt file")
ap.add_argument("-m", "--model", default="C:/study folder/IP project/facial-landmarks-master/res10_300x300_ssd_iter_140000.caffemodel",
help="Pre-trained caffe model")
ap.add_argument("-t", "--thresold", type=float, default=0.6,
help="Thresold value to filter weak detections")
args = ap.parse_args()
# whether it's hog or cnn or dl
model = args.learning.lower()
if model == 'hog':
# initialize hog + svm based face detector
face_detector = dlib.get_frontal_face_detector()
elif model == 'cnn':
# initialize cnn based face detector with the weights
face_detector = dlib.cnn_face_detection_model_v1(args.data)
if model == 'dl':
# Pre-trained caffe deep learning face detection model (SSD)
face_detector = cv2.dnn.readNetFromCaffe(args.prototxt, args.model)
else:
print("Please provide valid model name like cnn or hog")
exit()
# landmark predictor
predictor = dlib.shape_predictor(args.weights)
# if image is valid or not
image= None
if args.image:
# load input image
img = os.path.join(HOME, args.image)
image = cv2.imread(img)
if image is None:
print("Please provide image ...")
else:
print("Face detection for image")
face_detection(image)
in this code when I give the path of a single image it gives me output when I give the path to the images folder then I get an error can someone help me with this error, also can someone tell me how to find the accuracy of the images. please it’s urgent
