-
Notifications
You must be signed in to change notification settings - Fork 15
/
facial_landmarks_realtime.py
151 lines (120 loc) · 5.56 KB
/
facial_landmarks_realtime.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
"""
Created on Mon Dec 03 11:15:45 2018
@author: keyur-r
python facial_landmarks_realtime.py -l <> -w <> -p <> -m <> -t <>
l -> hog or dl
w -> model path for facial landmarks (shape_predictor_68_face_landmarks.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
from image_utility import save_image, generate_random_color, draw_border
from imutils import face_utils, video
def hog_detector(image, gray):
# Finding height and width of frame
(img_h, img_w) = image.shape[:2]
cv2.putText(image, "HOG method", (img_w - 200, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.5,
(205, 92, 92), 2)
# Get faces into webcam's image
rects = detector(gray, 0)
# For each detected face, find the landmark.
for (i, rect) in enumerate(rects):
# Finding points for rectangle to draw on face
x1, y1, x2, y2, w, h = rect.left(), rect.top(), rect.right() + \
1, rect.bottom() + 1, rect.width(), rect.height()
# cv2.rectangle(image, (x1, y1), (x1 + w, y1 + h), (205, 92, 92), 2)
draw_border(image, (x1, y1), (x2, y2), (205, 92, 92), 2, 10, 20)
# show the face number
cv2.putText(image, "Found #{}".format(i + 1), (x1 - 20, y1 - 20),
cv2.FONT_HERSHEY_TRIPLEX, 0.6, (205, 92, 92), 2)
# Make the prediction and transfom it to numpy array
shape = predictor(gray, 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_detector(image, gray):
# Facial landmarks with DL
# Finding height and width of frame
(h, w) = image.shape[:2]
cv2.putText(image, "DL method", (w - 200, 20), cv2.FONT_HERSHEY_SIMPLEX, 0.5,
(205, 92, 92), 2)
total_faces = 0
# 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))
detector.setInput(inputBlob)
detections = detector.forward()
for i in range(0, detections.shape[2]):
# Probability of prediction
prediction_score = detections[0, 0, i, 2]
if prediction_score < T:
continue
total_faces += 1
# 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), (205, 92, 92), 2)
draw_border(image, (x1, y1), (x2, y2), (205, 92, 92), 2, 10, 20)
# show the face number with confidence score
prediction_score_str = "{:.2f}%".format(prediction_score * 100)
label = "Found #{} ({})".format(total_faces, prediction_score_str)
cv2.putText(image, label, (x1 - 20, y1 - 20),
cv2.FONT_HERSHEY_TRIPLEX, 0.6, (205, 92, 92), 2)
# Draw on our image, all the finded cordinate points (x,y)
for (x, y) in shape:
cv2.circle(image, (x, y), 2, (0, 0, 255), -1)
def facial_landmarks():
# Feed from computer camera with threading
cap = video.VideoStream(src=0).start()
while True:
# Getting out image by webcam
image = cap.read()
# Converting the image to gray scale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
if face_detection_method == 'hog':
hog_detector(image, gray)
else:
dl_detector(image, gray)
# show the output frame
# adding brightness and contrast -> α⋅p(i,j)+β where p(i.j) is pixel value for each point
image = cv2.convertScaleAbs(image, alpha=1.0, beta=0)
cv2.imshow("Facial Landmarks", image)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
cv2.destroyAllWindows()
cap.stop()
if __name__ == "__main__":
# handle command line arguments
ap = argparse.ArgumentParser()
ap.add_argument('-w', '--weights',
default='./shape_predictor_68_face_landmarks.dat', help='Path to weights file')
ap.add_argument("-p", "--prototxt", default="../face_detection/deploy.prototxt.txt",
help="Caffe 'deploy' prototxt file")
ap.add_argument("-m", "--model", default="../face_detection/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")
ap.add_argument("-l", "--learning", default="hog",
help="Learning model from hog/dl")
args = ap.parse_args()
predictor = dlib.shape_predictor(args.weights)
face_detection_method = args.learning
T = args.thresold
if face_detection_method == 'hog':
detector = dlib.get_frontal_face_detector()
elif face_detection_method == 'dl':
detector = cv2.dnn.readNetFromCaffe(args.prototxt, args.model)
else:
print("Please select method from dl or hog to find landmarks")
facial_landmarks()