-
Notifications
You must be signed in to change notification settings - Fork 0
/
demo_onnx.py
153 lines (120 loc) · 4.03 KB
/
demo_onnx.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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import copy
import time
import argparse
import cv2 as cv
import numpy as np
import onnxruntime
def sigmoid(x: np.ndarray) -> np.ndarray:
return 1 / (1 + np.exp(-x))
def image_normalization(
image: np.ndarray,
image_min: int = 0,
image_max: int = 255,
epsilon: float = 1e-12,
) -> np.ndarray:
image = np.float32(image)
image = (image - np.min(image)) * (image_max - image_min) / (
(np.max(image) - np.min(image)) + epsilon) + image_min
return image
def run_inference(
onnx_session: onnxruntime.InferenceSession,
image: np.ndarray,
) -> np.ndarray:
# ONNX Input Size
input_size = onnx_session.get_inputs()[0].shape
input_width = input_size[3]
input_height = input_size[2]
# Pre process:Resize, BGR->RGB, Transpose, float32 cast
input_image = cv.resize(image, dsize=(input_width, input_height))
input_image = cv.cvtColor(input_image, cv.COLOR_BGR2RGB)
input_image = input_image.transpose(2, 0, 1)
input_image = np.expand_dims(input_image, axis=0)
input_image = input_image.astype('float32')
# Inference
input_name = onnx_session.get_inputs()[0].name
results = onnx_session.run(None, {input_name: input_image})
# Post process
image_width, image_height = image.shape[1], image.shape[0]
for index, result in enumerate(results):
temp = np.squeeze(result)
temp = sigmoid(temp)
temp = image_normalization(temp)
temp = temp.astype(np.uint8)
temp = cv.bitwise_not(temp)
temp = cv.resize(temp, dsize=(image_width, image_height))
results[index] = temp
average_image = np.uint8(np.mean(results, axis=0))
fuse_image = copy.deepcopy(results[index])
return average_image, fuse_image
def main():
parser = argparse.ArgumentParser()
parser.add_argument('--device', type=int, default=0)
parser.add_argument('--movie', type=str, default=None)
parser.add_argument('--image', type=str, default=None)
parser.add_argument(
'--model',
type=str,
default='model/LDC_640x360.onnx',
)
args = parser.parse_args()
model_path = args.model
image_path = args.image
# Initialize video capture
cap = None
if image_path is None:
cap_device = args.device
if args.movie is not None:
cap_device = args.movie
cap = cv.VideoCapture(cap_device)
# Load model
onnx_session = onnxruntime.InferenceSession(
model_path,
providers=[
'CUDAExecutionProvider',
'CPUExecutionProvider',
],
)
while True and image_path is None:
start_time = time.time()
# Capture read
ret, frame = cap.read()
if not ret:
break
debug_image = copy.deepcopy(frame)
# Inference execution
average_image, fuse_image = run_inference(
onnx_session,
frame,
)
elapsed_time = time.time() - start_time
# Draw Inference elapsed time
cv.putText(
debug_image,
"Elapsed Time : " + '{:.1f}'.format(elapsed_time * 1000) + "ms",
(10, 40), cv.FONT_HERSHEY_SIMPLEX, 1.0, (0, 255, 0), 2, cv.LINE_AA)
key = cv.waitKey(1)
if key == 27: # ESC
break
cv.imshow('LDC Input', debug_image)
cv.imshow('LDC Output(Average)', average_image)
cv.imshow('LDC Output(Fuse)', fuse_image)
if image_path is not None:
start_time = time.time()
# Read image
image = cv.imread(image_path)
# Inference execution
average_image, fuse_image = run_inference(
onnx_session,
image,
)
elapsed_time = time.time() - start_time
print("Elapsed Time : " + '{:.1f}'.format(elapsed_time * 1000) + "ms")
cv.imwrite('average_image.png', average_image)
cv.imwrite('fuse_image.png', fuse_image)
if cap is not None:
cap.release()
cv.destroyAllWindows()
if __name__ == '__main__':
main()