-
Notifications
You must be signed in to change notification settings - Fork 35
/
clean_alignfaces.py
309 lines (239 loc) · 11.3 KB
/
clean_alignfaces.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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
'''
AAA lllllll lllllll iiii
A:::A l:::::l l:::::l i::::i
A:::::A l:::::l l:::::l iiii
A:::::::A l:::::l l:::::l
A:::::::::A l::::l l::::l iiiiiii eeeeeeeeeeee
A:::::A:::::A l::::l l::::l i:::::i ee::::::::::::ee
A:::::A A:::::A l::::l l::::l i::::i e::::::eeeee:::::ee
A:::::A A:::::A l::::l l::::l i::::i e::::::e e:::::e
A:::::A A:::::A l::::l l::::l i::::i e:::::::eeeee::::::e
A:::::AAAAAAAAA:::::A l::::l l::::l i::::i e:::::::::::::::::e
A:::::::::::::::::::::A l::::l l::::l i::::i e::::::eeeeeeeeeee
A:::::AAAAAAAAAAAAA:::::A l::::l l::::l i::::i e:::::::e
A:::::A A:::::A l::::::ll::::::li::::::ie::::::::e
A:::::A A:::::A l::::::ll::::::li::::::i e::::::::eeeeeeee
A:::::A A:::::A l::::::ll::::::li::::::i ee:::::::::::::e
AAAAAAA AAAAAAAlllllllllllllllliiiiiiii eeeeeeeeeeeeee
_____ _ _ ___ ______ _____
/ __ \ | (_) / _ \ | ___ \_ _| _
| / \/ | ___ __ _ _ __ _ _ __ __ _ / /_\ \| |_/ / | | (_)
| | | |/ _ \/ _` | '_ \| | '_ \ / _` | | _ || __/ | |
| \__/\ | __/ (_| | | | | | | | | (_| | | | | || | _| |_ _
\____/_|\___|\__,_|_| |_|_|_| |_|\__, | \_| |_/\_| \___/ (_)
__/ |
|___/
_ _ _ _
| | | (_) | |
| | | |_ __| | ___ ___
| | | | |/ _` |/ _ \/ _ \
\ \_/ / | (_| | __/ (_) |
\___/|_|\__,_|\___|\___/
This script takes a folder of .mp4 video files and converts
them to a folder of .png files full of faces (centered and aligned).
The original video files are also placed in these folders.
This is enabled if default_video_cleaners=['clean_alignfaces']
'''
# you only use these modules if you register, so put them here
import cv2, os, time, shutil, math
import skvideo.io, skvideo.motion, skvideo.measure
from moviepy.editor import VideoFileClip
from PIL import Image
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from PIL import Image
def clean_alignfaces(videofile, basedir):
def euclidean_distance(a, b):
x1 = a[0]; y1 = a[1]
x2 = b[0]; y2 = b[1]
return math.sqrt(((x2 - x1) * (x2 - x1)) + ((y2 - y1) * (y2 - y1)))
def detectFace(img):
faces = face_detector.detectMultiScale(img, 1.3, 5)
#print("found faces: ", len(faces))
if len(faces) > 0:
face = faces[0]
face_x, face_y, face_w, face_h = face
img = img[int(face_y):int(face_y+face_h), int(face_x):int(face_x+face_w)]
img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
return img, img_gray
else:
img_gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
return img, img_gray
#raise ValueError("No face found in the passed image ")
def alignFace(img_path):
img = cv2.imread(img_path)
# plt.imshow(img[:, :, ::-1])
# plt.show()
img_raw = img.copy()
img, gray_img = detectFace(img)
eyes = eye_detector.detectMultiScale(gray_img)
#print("found eyes: ",len(eyes))
if len(eyes) >= 2:
#find the largest 2 eye
base_eyes = eyes[:, 2]
#print(base_eyes)
items = []
for i in range(0, len(base_eyes)):
item = (base_eyes[i], i)
items.append(item)
df = pd.DataFrame(items, columns = ["length", "idx"]).sort_values(by=['length'], ascending=False)
eyes = eyes[df.idx.values[0:2]]
#--------------------
#decide left and right eye
eye_1 = eyes[0]; eye_2 = eyes[1]
if eye_1[0] < eye_2[0]:
left_eye = eye_1
right_eye = eye_2
else:
left_eye = eye_2
right_eye = eye_1
#--------------------
#center of eyes
left_eye_center = (int(left_eye[0] + (left_eye[2] / 2)), int(left_eye[1] + (left_eye[3] / 2)))
left_eye_x = left_eye_center[0]; left_eye_y = left_eye_center[1]
right_eye_center = (int(right_eye[0] + (right_eye[2]/2)), int(right_eye[1] + (right_eye[3]/2)))
right_eye_x = right_eye_center[0]; right_eye_y = right_eye_center[1]
#center_of_eyes = (int((left_eye_x+right_eye_x)/2), int((left_eye_y+right_eye_y)/2))
cv2.circle(img, left_eye_center, 2, (255, 0, 0) , 2)
cv2.circle(img, right_eye_center, 2, (255, 0, 0) , 2)
#cv2.circle(img, center_of_eyes, 2, (255, 0, 0) , 2)
#----------------------
#find rotation direction
if left_eye_y > right_eye_y:
point_3rd = (right_eye_x, left_eye_y)
direction = -1 #rotate same direction to clock
print("rotate to clock direction")
else:
point_3rd = (left_eye_x, right_eye_y)
direction = 1 #rotate inverse direction of clock
print("rotate to inverse clock direction")
#----------------------
cv2.circle(img, point_3rd, 2, (255, 0, 0) , 2)
cv2.line(img,right_eye_center, left_eye_center,(67,67,67),1)
cv2.line(img,left_eye_center, point_3rd,(67,67,67),1)
cv2.line(img,right_eye_center, point_3rd,(67,67,67),1)
a = euclidean_distance(left_eye_center, point_3rd)
b = euclidean_distance(right_eye_center, point_3rd)
c = euclidean_distance(right_eye_center, left_eye_center)
#print("left eye: ", left_eye_center)
#print("right eye: ", right_eye_center)
#print("additional point: ", point_3rd)
#print("triangle lengths: ",a, b, c)
cos_a = (b*b + c*c - a*a)/(2*b*c)
#print("cos(a) = ", cos_a)
angle = np.arccos(cos_a)
#print("angle: ", angle," in radian")
angle = (angle * 180) / math.pi
print("angle: ", angle," in degree")
if direction == -1:
angle = 90 - angle
print("angle: ", angle," in degree")
#--------------------
#rotate image
new_img = Image.fromarray(img_raw)
new_img = np.array(new_img.rotate(direction * angle))
else:
#find the largest 2 ey
new_img = img_raw
return new_img
def capture_video(filename, timesplit):
video=cv2.VideoCapture(0)
fourcc = cv2.VideoWriter_fourcc(*'XVID')
frame_width = int(video.get(3))
frame_height = int(video.get(4))
out = cv2.VideoWriter(filename,cv2.VideoWriter_fourcc('M','J','P','G'), 10, (frame_width,frame_height))
a=0
start=time.time()
while True:
a=a+1
check, frame=video.read()
#print(check)
#print(frame)
gray=cv2.cvtColor(frame,cv2.COLOR_BGR2GRAY)
out.write(frame)
#cv2.imshow("frame",gray)
end=time.time()
if end-start>timesplit:
break
#print(end-start)
print(a)
video.release()
out.release()
cv2.destroyAllWindows()
return filename
def cut_faces(modeldir,filename):
# import data later
hostdir=os.getcwd()
# capture_video(filename, 5)
face_cascade = cv2.CascadeClassifier(os.getcwd()+'/helpers/haarcascade_frontalface_default.xml')
foldername=filename[0:-4]+'_faces'
try:
os.mkdir(foldername)
except:
shutil.rmtree(foldername)
os.mkdir(foldername)
shutil.copy(hostdir+'/'+filename, hostdir+'/'+foldername+'/'+filename)
os.chdir(foldername)
videodata=skvideo.io.vread(filename)
frames, rows, cols, channels = videodata.shape
metadata=skvideo.io.ffprobe(filename)
frame=videodata[0]
r,c,ch=frame.shape
for i in range(0,len(videodata),25):
#row, col, channels
skvideo.io.vwrite("output"+str(i)+".png", videodata[i])
listdir=os.listdir()
facenums=0
for i in range(len(listdir)):
if listdir[i][-4:]=='.png':
try:
image_file = listdir[i]
alignedFace = alignFace(image_file)
gray = cv2.cvtColor(alignedFace, cv2.COLOR_BGR2GRAY)
faces = face_cascade.detectMultiScale(gray, 1.3, 5)
increment=0
print(len(faces))
files=list()
if len(faces) == 0:
pass
else:
for (x,y,w,h) in faces:
img=alignedFace
os.chdir(hostdir+'/'+foldername)
cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
newimg=img[y:y+h,x:x+w]
new_image_file=image_file[0:-4] + '_face_' + str(increment) + '.png'
newimg=cv2.resize(newimg, (100, 100), interpolation=cv2.INTER_LINEAR)
norm_img = np.zeros((100, 100))
norm_img = cv2.normalize(newimg, norm_img, 0, 255, cv2.NORM_MINMAX)
cv2.imwrite(new_image_file, newimg)
files.append(new_image_file)
facenums=facenums+1
except:
print('error')
os.chdir(hostdir+'/'+foldername)
listdir=os.listdir()
print(listdir)
for i in range(len(listdir)):
if listdir[i][-4:]=='.png':
if listdir[i].find('face') < 0:
os.remove(listdir[i])
return facenums, files
# paths
opencv_home = cv2.__file__
folders = opencv_home.split(os.path.sep)[0:-1]
path = folders[0]
for folder in folders[1:]:
path = path + "/" + folder
# other stuff
face_detector_path = basedir+"/helpers/haarcascade_frontalface_default.xml"
eye_detector_path = basedir+"/helpers/haarcascade_eye.xml"
nose_detector_path = basedir+"/helpers/haarcascade_mcs_nose.xml"
if os.path.isfile(face_detector_path) != True:
raise ValueError("Confirm that opencv is installed on your environment! Expected path ",detector_path," violated.")
face_detector = cv2.CascadeClassifier(face_detector_path)
eye_detector = cv2.CascadeClassifier(eye_detector_path)
nose_detector = cv2.CascadeClassifier(nose_detector_path)
facenums, files=cut_faces(os.getcwd(), videofile)
return [filename]