-
Notifications
You must be signed in to change notification settings - Fork 26
/
mtcnn_tfv2.py
56 lines (46 loc) · 1.73 KB
/
mtcnn_tfv2.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
import argparse
import tensorflow as tf
import cv2
def mtcnn_fun(img, min_size, factor, thresholds):
with open('./mtcnn.pb', 'rb') as f:
graph_def = tf.compat.v1.GraphDef.FromString(f.read())
with tf.device('/cpu:0'):
prob, landmarks, box = tf.compat.v1.import_graph_def(graph_def,
input_map={
'input:0': img,
'min_size:0': min_size,
'thresholds:0': thresholds,
'factor:0': factor
},
return_elements=[
'prob:0',
'landmarks:0',
'box:0']
, name='')
print(box, prob, landmarks)
return box, prob, landmarks
# wrap graph function as a callable function
mtcnn_fun = tf.compat.v1.wrap_function(mtcnn_fun, [
tf.TensorSpec(shape=[None, None, 3], dtype=tf.float32),
tf.TensorSpec(shape=[], dtype=tf.float32),
tf.TensorSpec(shape=[], dtype=tf.float32),
tf.TensorSpec(shape=[3], dtype=tf.float32)
])
def main(args):
img = cv2.imread(args.image)
bbox, scores, landmarks = mtcnn_fun(img, 40, 0.7, [0.6, 0.7, 0.8])
bbox, scores, landmarks = bbox.numpy(), scores.numpy(), landmarks.numpy()
print('total box:', len(bbox))
for box, pts in zip(bbox, landmarks):
box = box.astype('int32')
img = cv2.rectangle(img, (box[1], box[0]), (box[3], box[2]), (255, 0, 0), 3)
pts = pts.astype('int32')
for i in range(5):
img = cv2.circle(img, (pts[i+5], pts[i]), 1, (0, 255, 0), 2)
cv2.imshow('image', img)
cv2.waitKey(0)
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='tensorflow mtcnn')
parser.add_argument('image', help='image path')
args = parser.parse_args()
main(args)