ReFace/faceswapper.py at main · Am1riki/ReFace · GitHub

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
import cv2
import numpy as np
import dlib
from imutils.video import VideoStream, FPS

def index_nparray(nparray):
    index = None
    for num in nparray[0]:
            index = num
            break
    return index

def get_points_list(img, detector, predictor):
    graymask = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) #maska bolatyn suretti sur tuske auystyru
    faces  = detector(graymask)
    points_lists = []
    for face in faces:
        x1 = face.left()
        y1 = face.top()
        x2 = face.right()
        y2 = face.bottom()
        landmarks = predictor(graymask, face)
        for n in range(0, 68):
            x = landmarks.part(n).x
            y = landmarks.part(n).y
            points_lists += [(x, y)]
    return points_lists

def deepFake(img, img2, detector, predictor):
    points_list  = get_points_list(img, detector, predictor)
    #print(points_list)
    points_list2 = get_points_list(img2, detector, predictor)
    graymask = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    points = np.array(points_list, np.int32)
    #print(points)
    convexhull = cv2.convexHull(points)
    mask = np.zeros_like(graymask)
    cv2.fillConvexPoly(mask, convexhull, 255)
    face_image_1 = cv2.bitwise_and(img, img, mask=mask)
    rect = cv2.boundingRect(convexhull)
    subdiv = cv2.Subdiv2D(rect)
    subdiv.insert(points_list)
    triangles = subdiv.getTriangleList()
    triangles = np.array(triangles, dtype=np.int32)
    triangles_id = []
    for t in triangles:
            pt1 = (t[0], t[1])
            pt2 = (t[2], t[3])
            pt3 = (t[4], t[5])
            id_pt1 = np.where((points == pt1).all(axis=1))
            id_pt1 = index_nparray(id_pt1)
            id_pt2 = np.where((points == pt2).all(axis=1))
            id_pt2 = index_nparray(id_pt2)
            id_pt3 = np.where((points == pt3).all(axis=1))
            id_pt3 = index_nparray(id_pt3)

            if id_pt1 is not None and id_pt2 is not None and id_pt3 is not None:
                triangle = [id_pt1, id_pt2, id_pt3]
                triangles_id.append(triangle)

    points2 = np.array(points_list2, np.int32)#
    #print(points2)
    convexhull2 = cv2.convexHull(points2)#
    img2_new_face = np.zeros_like(img2, np.uint8)
    for triangle_index in triangles_id:
            tr1_pt1 = points_list[triangle_index[0]]
            tr1_pt2 = points_list[triangle_index[1]]
            tr1_pt3 = points_list[triangle_index[2]]
            triangle1 = np.array([tr1_pt1, tr1_pt2, tr1_pt3], np.int32)
            rect1 = cv2.boundingRect(triangle1)
            (x1, y1, w1, h1) = rect1
            cropped_triangle = img[y1: y1 + h1, x1: x1 + w1]
            cropped_tr1_mask = np.zeros((h1, w1), np.uint8)
            points = np.array([[tr1_pt1[0] - x1, tr1_pt1[1] - y1],
                      [tr1_pt2[0] - x1, tr1_pt2[1] - y1],
                      [tr1_pt3[0] - x1, tr1_pt3[1] - y1]], np.int32)
            cv2.fillConvexPoly(cropped_tr1_mask, points, 255)
            cropped_triangle = cv2.bitwise_and(cropped_triangle, cropped_triangle, mask=cropped_tr1_mask)

            tr2_pt1 = points_list2[triangle_index[0]]
            tr2_pt2 = points_list2[triangle_index[1]]
            tr2_pt3 = points_list2[triangle_index[2]]
            triangle2 = np.array([tr2_pt1, tr2_pt2, tr2_pt3], np.int32)
            rect2 = cv2.boundingRect(triangle2)
            (x2, y2, w2, h2) = rect2
            cropped_triangle2 = img2[y2: y2 + h2, x2: x2 + w2]
            cropped_tr2_mask = np.zeros((h2, w2), np.uint8)
            points2 = np.array([[tr2_pt1[0] - x2, tr2_pt1[1] - y2],
                       [tr2_pt2[0] - x2, tr2_pt2[1] - y2],
                       [tr2_pt3[0] - x2, tr2_pt3[1] - y2]], np.int32)
            cv2.fillConvexPoly(cropped_tr2_mask, points2, 255)
            cropped_triangle2 = cv2.bitwise_and(cropped_triangle2, cropped_triangle2, mask=cropped_tr2_mask)

            points = np.float32(points)
            points2 = np.float32(points2)
            M = cv2.getAffineTransform(points, points2)
            warped_triangle = cv2.warpAffine(cropped_triangle, M, (w2, h2),flags=cv2.INTER_NEAREST)
            warped_triangle = cv2.bitwise_and(warped_triangle, warped_triangle, mask=cropped_tr2_mask)


            img2_new_face_rect_area = img2_new_face[y2: y2 + h2, x2: x2 + w2]
            img2_new_face_rect_area_gray = cv2.cvtColor(img2_new_face_rect_area, cv2.COLOR_BGR2GRAY)

            _, mask_triangles_designed = cv2.threshold(img2_new_face_rect_area_gray, 1, 255, cv2.THRESH_BINARY_INV)
            warped_triangle = cv2.bitwise_and(warped_triangle, warped_triangle, mask=mask_triangles_designed)

            img2_new_face_rect_area = cv2.add(img2_new_face_rect_area, warped_triangle)
            img2_new_face[y2: y2 + h2, x2: x2 + w2] = img2_new_face_rect_area

    img2_face_mask = np.zeros_like(cv2.cvtColor(img2, cv2.COLOR_BGR2GRAY))
    img2_head_mask = cv2.fillConvexPoly(img2_face_mask, convexhull2, 255)
    img2_face_mask = cv2.bitwise_not(img2_head_mask)
    img2_noface = cv2.bitwise_and(img2, img2, mask=img2_face_mask)
    result = cv2.add(img2_noface, img2_new_face)

    (x3, y3, w3, h3) = cv2.boundingRect(convexhull2)
    center_face = (int((x3 + x3 + w3) / 2), int((y3 + y3 + h3) / 2))

    img2_face_mask = np.zeros_like(cv2.cvtColor(img2, cv2.COLOR_BGR2GRAY))
    img2_head_mask = cv2.fillConvexPoly(img2_face_mask, convexhull2, 255)
    img2_face_mask = cv2.bitwise_not(img2_head_mask)
    img2_noface = cv2.bitwise_and(img2, img2, mask=img2_face_mask)
    result = cv2.add(img2_noface, img2_new_face)

    (x3, y3, w3, h3) = cv2.boundingRect(convexhull2)
    center_face = (int((x3 + x3 + w3) / 2), int((y3 + y3 + h3) / 2))
    seamlessclone = cv2.seamlessClone(result, img2, img2_head_mask, center_face, cv2.MONOCHROME_TRANSFER)
    return seamlessclone


def savevideo(imgpath, videopath):
    detector = dlib.get_frontal_face_detector()
    predictor = dlib.shape_predictor('C:/Users/am1ri/Desktop/DeepFake/shape_predictor_68_face_landmarks.dat/shape_predictor_68_face_landmarks.dat')
    img = cv2.imread(imgpath)
    cap = cv2.VideoCapture(videopath)
    if (cap.isOpened() == False):
        print("Error reading video file")
    frame_width = int(cap.get(3))
    frame_height = int(cap.get(4))
    video_cod = cv2.VideoWriter_fourcc(*'XVID')
    #video_cod = cv2.VideoWriter_fourcc(*'mp4v')
    path = 'C:/Users/am1ri/Desktop/kivy project/output.avi'
    video_output = cv2.VideoWriter(path, video_cod, 30,(frame_width,frame_height))
    while True:
            ret,frame = cap.read()
            if ret == True:
                try:
                    deeps  = deepFake(img, frame, detector, predictor)
                    video_output.write(deeps)
                except:
                    pass
            else:
                break

    cap.release()
    return path


photo = 'C:\\Users\\am1ri\\Desktop\\DeepFake\\Media\\evans.jpg'
video = 'C:\\Users\\am1ri\\Desktop\\DeepFake\\Media\\Sups5.mp4'
print(savevideo(photo, video))