forked from serengil/tensorflow-101
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdeep-face-real-time.py
159 lines (121 loc) · 5.86 KB
/
deep-face-real-time.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
152
153
154
155
156
157
158
159
#author Sefik Ilkin Serengil
#you can find the documentation of this code from the following link: https://sefiks.com/2018/08/06/deep-face-recognition-with-keras/
import numpy as np
import cv2
from keras.models import Model, Sequential
from keras.layers import Input, Convolution2D, ZeroPadding2D, MaxPooling2D, Flatten, Dense, Dropout, Activation
from PIL import Image
from keras.preprocessing.image import load_img, save_img, img_to_array
from keras.applications.imagenet_utils import preprocess_input
from keras.preprocessing import image
import matplotlib.pyplot as plt
from os import listdir
#-----------------------
color = (67,67,67)
face_cascade = cv2.CascadeClassifier('C:/Users/IS96273/AppData\Local/Continuum/anaconda3/pkgs/opencv-3.3.1-py35h20b85fd_1/Library/etc/haarcascades/haarcascade_frontalface_default.xml')
def preprocess_image(image_path):
img = load_img(image_path, target_size=(224, 224))
img = img_to_array(img)
img = np.expand_dims(img, axis=0)
#preprocess_input normalizes input in scale of [-1, +1]. You must apply same normalization in prediction.
#Ref: https://github.com/keras-team/keras-applications/blob/master/keras_applications/imagenet_utils.py (Line 45)
img = preprocess_input(img)
return img
def loadVggFaceModel():
model = Sequential()
model.add(ZeroPadding2D((1,1),input_shape=(224,224, 3)))
model.add(Convolution2D(64, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D((2,2), strides=(2,2)))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(128, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(128, (3, 3), activation='relu'))
model.add(MaxPooling2D((2,2), strides=(2,2)))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(256, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(256, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(256, (3, 3), activation='relu'))
model.add(MaxPooling2D((2,2), strides=(2,2)))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(512, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(512, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(512, (3, 3), activation='relu'))
model.add(MaxPooling2D((2,2), strides=(2,2)))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(512, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(512, (3, 3), activation='relu'))
model.add(ZeroPadding2D((1,1)))
model.add(Convolution2D(512, (3, 3), activation='relu'))
model.add(MaxPooling2D((2,2), strides=(2,2)))
model.add(Convolution2D(4096, (7, 7), activation='relu'))
model.add(Dropout(0.5))
model.add(Convolution2D(4096, (1, 1), activation='relu'))
model.add(Dropout(0.5))
model.add(Convolution2D(2622, (1, 1)))
model.add(Flatten())
model.add(Activation('softmax'))
#you can download pretrained weights from https://drive.google.com/file/d/1CPSeum3HpopfomUEK1gybeuIVoeJT_Eo/view?usp=sharing
from keras.models import model_from_json
model.load_weights('C:/Users/IS96273/Desktop/vgg_face_weights.h5')
vgg_face_descriptor = Model(inputs=model.layers[0].input, outputs=model.layers[-2].output)
return vgg_face_descriptor
model = loadVggFaceModel()
#------------------------
#put your employee pictures in this path as name_of_employee.jpg
employee_pictures = "C:/Users/IS96273/Desktop/database/"
employees = dict()
for file in listdir(employee_pictures):
employee, extension = file.split(".")
employees[employee] = model.predict(preprocess_image('C:/Users/IS96273/Desktop/database/%s.jpg' % (employee)))[0,:]
print("employee representations retrieved successfully")
def findCosineSimilarity(source_representation, test_representation):
a = np.matmul(np.transpose(source_representation), test_representation)
b = np.sum(np.multiply(source_representation, source_representation))
c = np.sum(np.multiply(test_representation, test_representation))
return 1 - (a / (np.sqrt(b) * np.sqrt(c)))
#------------------------
cap = cv2.VideoCapture(0) #webcam
#cap = cv2.VideoCapture('C:/Users/IS96273/Desktop/zuckerberg.mp4') #video
while(True):
ret, img = cap.read()
#img = cv2.resize(img, (640, 360))
faces = face_cascade.detectMultiScale(img, 1.3, 5)
for (x,y,w,h) in faces:
if w > 130:
#cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2) #draw rectangle to main image
detected_face = img[int(y):int(y+h), int(x):int(x+w)] #crop detected face
detected_face = cv2.resize(detected_face, (224, 224)) #resize to 224x224
img_pixels = image.img_to_array(detected_face)
img_pixels = np.expand_dims(img_pixels, axis = 0)
#img_pixels /= 255
#employee dictionary is using preprocess_image and it normalizes in scale of [-1, +1]
img_pixels /= 127.5
img_pixels -= 1
captured_representation = model.predict(img_pixels)[0,:]
found = 0
for i in employees:
employee_name = i
representation = employees[i]
similarity = findCosineSimilarity(representation, captured_representation)
if(similarity < 0.30):
cv2.putText(img, employee_name, (int(x+w+15), int(y-12)), cv2.FONT_HERSHEY_SIMPLEX, 1, color, 2)
found = 1
break
#connect face and text
cv2.line(img,(int((x+x+w)/2),y+15),(x+w,y-20),color,1)
cv2.line(img,(x+w,y-20),(x+w+10,y-20),color,1)
if(found == 0): #if found image is not in employee database
cv2.putText(img, 'unknown', (int(x+w+15), int(y-12)), cv2.FONT_HERSHEY_SIMPLEX, 1, color, 2)
cv2.imshow('img',img)
if cv2.waitKey(1) & 0xFF == ord('q'): #press q to quit
break
#kill open cv things
cap.release()
cv2.destroyAllWindows()