-
Notifications
You must be signed in to change notification settings - Fork 2
/
tf-trt-ssd.py
142 lines (119 loc) · 4.92 KB
/
tf-trt-ssd.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
# MIT License
#
# Copyright (c) 2019 Tsutomu Furuse
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in all
# copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.
from __future__ import print_function
import cv2
import tensorflow.contrib.tensorrt as trt
import tensorflow as tf
import numpy as np
import ast
FRAME_WIDTH = 1280
FRAME_HEIGHT = 960
GST_STR = 'nvarguscamerasrc \
! video/x-raw(memory:NVMM), width=3280, height=2464, format=(string)NV12, framerate=(fraction)30/1 \
! nvvidconv ! video/x-raw, width=(int)%d, height=(int)%d, format=(string)BGRx \
! videoconvert \
! appsink' % (FRAME_WIDTH, FRAME_HEIGHT)
WINDOW_NAME = 'TF-TRT Object Detection'
MODEL_FILE = './ssd_inception_v2_coco_trt.pb'
LABEL_FILE = './coco-labels-paper.txt'
def load_graph_def(file):
with tf.gfile.GFile(file, 'rb') as f:
graph_def = tf.GraphDef()
graph_def.ParseFromString(f.read())
return graph_def
def load_labels(file):
labels = ['unlabeled']
with open(file, 'r') as f:
for line in f.read().splitlines():
labels.append(line)
return labels
def main():
'''
A simple python application to detect objects from camera captured image
using TF-TRT for NVIDIA Jetson Nano Developer Kit.
This application assumes the TensorRT optimized ssd_mobilenet_v1_coco model.
Refer to the NVIDIA-AI-IOT/tf_trt_models GitHub ripository for details on
the model.
'''
labels = load_labels(LABEL_FILE)
num_labels = len(labels)
print('Loading graph definition...', end = '', flush = True)
trt_graph_def = load_graph_def(MODEL_FILE)
print('Done.')
tf_config = tf.ConfigProto()
tf_config.gpu_options.allow_growth = True
tf_sess = tf.Session(config = tf_config)
print('Importing graph definition to TensorFlow...', \
end = '', flush = True)
tf.import_graph_def(trt_graph_def, name = '')
print('Done.')
input_names = ['image_tensor']
tf_input = tf_sess.graph.get_tensor_by_name(input_names[0] + ':0')
tf_scores = tf_sess.graph.get_tensor_by_name('detection_scores:0')
tf_boxes = tf_sess.graph.get_tensor_by_name('detection_boxes:0')
tf_classes = tf_sess.graph.get_tensor_by_name('detection_classes:0')
tf_num_detections = tf_sess.graph.get_tensor_by_name('num_detections:0')
print('Configuring camera...', end = '', flush = True)
cap = cv2.VideoCapture(GST_STR, cv2.CAP_GSTREAMER)
print('Done.')
while True:
ret, img = cap.read()
if ret != True:
break
# Caputure frame
imgConv = cv2.cvtColor(img, cv2.COLOR_BGRA2BGR)
imgRs = cv2.resize(imgConv, (300, 300))
# Do inference
scores, boxes, classes, num_detections \
= tf_sess.run( \
[tf_scores, tf_boxes, tf_classes, tf_num_detections], \
feed_dict={tf_input: imgRs[None, ...]})
boxes = boxes[0] # index by 0 to remove batch dimension
scores = scores[0]
classes = classes[0]
num_detections = num_detections[0]
for i in range(int(num_detections)):
# Look up label string
class_id = int(classes[i])
label = labels[class_id] if class_id < num_labels else 'unlabeled'
# Get score
score = scores[i]
# Draw bounding box
box = boxes[i] * np.array( \
[FRAME_HEIGHT, FRAME_WIDTH, FRAME_HEIGHT, FRAME_WIDTH])
box = box.astype(np.int)
cv2.rectangle(img, \
(box[1], box[0]), (box[3], box[2]), (0, 255, 0), 3)
# Put label near bounding box
inf = '%s: %f' % (label, score)
print(inf)
cv2.putText(img, inf, (box[1], box[2]), \
cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 255, 0), 1, cv2.LINE_AA)
# Show image
cv2.imshow(WINDOW_NAME, img)
# Check if user hits ESC key to exit
key = cv2.waitKey(1)
if key == 27: # ESC
break
tf_sess.close()
if __name__ == "__main__":
main()