Skip to content

Commit 8bbb08a

Browse files
ivanvikhrevivanvikhrev
committed
add readme
1 parent f0ff4a4 commit 8bbb08a

19 files changed

+411
-261
lines changed

‎demos/python_demos/smart_library_demo/README.md

+131
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,131 @@
1+
# Smart Library Demo
2+
The applicattion is a demo of automated “smart library”.
3+
It involves the registration of the reader; authorization of the reader through face recognition;
4+
receiving and returning books by recognizing QR codes generated for each book in the library.
5+
The following pretrained models can be used:
6+
7+
* `face-detection-retail-0004`, to detect faces and predict their bounding boxes;
8+
* `landmarks-regression-retail-0009`, to predict face keypoints;
9+
* `face-reidentification-retail-0095`, to recognize readers.
10+
11+
For more information about the pre-trained models, refer to the [model documentation](../../../models/intel/index.md).
12+
13+
### How it works
14+
15+
The application is started from command line. It accepted several parameters.
16+
It reads video stream frame-by-frame from a web-camera device and performs independent analysis
17+
of each frame. To make predictions the application using 3 models. An input frame is processed by
18+
the face detection model to predict face bounding boxes. Then, face keypoints
19+
are predicted by the facial landmarks regression model. Keypoints are using
20+
to align the face and match it with face in data base.
21+
To register in library press `r` on keyboard. Once reader has beeen registered, he will be autothorized
22+
through face recognition. Registration allows to receiving and ruturning books by recognizing QR codes for
23+
each book. To make book recognition press `b` on keyboard. Also applications provides some
24+
extra statictics in console, like list of registered readers, full list of books in the library,
25+
hystory of borrowing books. To change information in console press `f` on keyboard. To exit press `q`.
26+
27+
### Creating QR-codes for books
28+
Next two files are using to create QR-codes for books:
29+
`library.json` file contains information about books in the library.
30+
`createQRCodes.py` script generates QR-codes for each book in `library.json` file.
31+
32+
usage: createQRCodes.py [-h] [-i LIB] [-o OUT]
33+
34+
optional arguments:
35+
-h, --help show this help message and exit
36+
-i LIB unput `.json` file with info
37+
-o OUT directory to save generated QR-codes
38+
39+
### Installation and dependencies
40+
41+
The demo depends on:
42+
- OpenVINO toolkit (2019R3 or newer)
43+
- Python (any of 2.7+ or 3.4+, which is supported by OpenVINO)
44+
- OpenCV (>=3.4.0)
45+
46+
To install all the required Python modules you can use:
47+
48+
pip install -r requirements.txt
49+
50+
51+
### Running the demo:
52+
53+
Running the application with the `-h` option or without
54+
any arguments yields the following message:
55+
56+
``` sh
57+
python ./smart_library_demo.py -h
58+
59+
usage: smart_library_demo.py [-h] -reid RDDET -m_rd RDMODEL -fd FDDET -m_fd
60+
FDMODEL -lm LMDET -m_lm LMMODEL [-w_rd RDWIDTH]
61+
[-h_rd RDHEIGHT] [-t_rd RDTHRESHOLD]
62+
[-w_fd FDWIDTH] [-h_fd FDHEIGHT]
63+
[-t_fd FDTHRESHOLD] [-w_lm LMWIDTH]
64+
[-h_lm LMHEIGHT] [-br BR] [-lib LIB] [-w WEB]
65+
66+
Smart Library Sample
67+
68+
Optional arguments:
69+
-h, --help show this help message and exit
70+
-w_rd RDWIDTH Optional. Image width to resize
71+
-h_rd RDHEIGHT Optional. Image height to resize
72+
-t_rd RDTHRESHOLD Optional. Probability threshold for face detections.
73+
-w_fd FDWIDTH Optional. Image width to resize
74+
-h_fd FDHEIGHT Optional. Image height to resize
75+
-t_fd FDTHRESHOLD Optional. Probability threshold for face detections.
76+
-w_lm LMWIDTH Optional. Image width to resize
77+
-h_lm LMHEIGHT Optional. Image height to resize
78+
-br BR Optional. Type - QR
79+
-lib LIB Optional. Path to library.
80+
-w WEB Optional. Specify index of web-camera to open. Default is
81+
0
82+
83+
Models:
84+
-reid RDDET Required. Type of recognizer. Available DNN face
85+
recognizer - DNNfr
86+
-m_rd RDMODEL Required. Path to .xml file
87+
-fd FDDET Required. Type of detector. Available DNN face detector -
88+
DNNfd
89+
-m_fd FDMODEL Required. Path to .xml file
90+
-lm LMDET Required. Type of detector. Available DNN landmarks
91+
regression - DNNlm
92+
-m_lm LMMODEL Required. Path to .xml file
93+
94+
95+
Example of a valid command line to run the application:
96+
97+
Linux (`sh`, `bash`, ...) (assuming OpenVINO installed in `/opt/intel/openvino`):
98+
99+
``` sh
100+
# Set up the environment
101+
source /opt/intel/openvino/bin/setupvars.sh
102+
103+
python ./smart_library_demo.py \
104+
-reid='DNNfr' \
105+
-m_rd=<path_to_model>/face-reidentification-retail-0095.xml \
106+
-fd='DNNfd' -m_fd=<path_to_model>/face-detection-retail-0004.xml \
107+
-lm='DNNlm' \
108+
-m_lm=<path_to_model>/landmarks-regression-retail-0009.xml \
109+
```
110+
111+
Windows (`cmd`, `powershell`) (assuming OpenVINO installed in `C:/Program Files (x86)/IntelSWTools/openvino/`):
112+
113+
``` powershell
114+
# Set up the environment
115+
call C:/Program Files (x86)/IntelSWTools/openvino_2019.3.334/bin/setupvars.bat
116+
117+
python smart_library_demo.py -reid='DNNfr' -m_rd=<path_to_model>/face-reidentification-retail-0095.xml
118+
-fd='DNNfd' -m_fd=<path_to_model>/face-detection-retail-0004.xml
119+
-lm='DNNlm' -m_lm=<path_to_model>/landmarks-regression-retail-0009.xml
120+
121+
Notice that the custom networks should be converted to the
122+
Inference Engine format (*.xml + *bin) first. To do this use the
123+
[Model Optimizer](https://software.intel.com/en-us/articles/OpenVINO-ModelOptimizer) tool.
124+
125+
### Demo output
126+
127+
The demo uses OpenCV window to display the resulting video frame and detections.
128+
It outputs logs to the terminal.
129+
130+
## See also
131+
* [Using Inference Engine Demos](../../README.md)

‎demos/python_demos/smart_library_demo/createQRCodes.py

+39
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,39 @@
1+
import sys, os
2+
import argparse
3+
import json
4+
sys.path.append("src/modules")
5+
import QR_generator as qr
6+
7+
8+
def createArgparse():
9+
parser = argparse.ArgumentParser()
10+
parser.add_argument('-i', type=str, dest = 'lib', default='library.json')
11+
parser.add_argument('-o', type=str, dest = 'out', default='qr-codes')
12+
return parser.parse_args()
13+
def main():
14+
args = createArgparse()
15+
gen = qr.QRgenerator()
16+
17+
18+
if (args.lib != None and os.path.isfile(args.lib) and args.out != None ):
19+
try:
20+
os.mkdir(args.out)
21+
except OSError:
22+
print ("Creation of the directory %s failed" % args.out)
23+
else:
24+
print ("Successfully created the directory %s " % args.out)
25+
with open(args.lib, 'r', encoding='utf-8') as lib:
26+
data = json.load(lib)
27+
28+
for book in data['books']:
29+
strData = (str(book['id'])+ ' ' + book['title'] + ' ' +
30+
book['author'] + ' ' + book['publisher'] + ' ' +
31+
str(book['year']))
32+
qr = gen.makeQR(strData)
33+
print(strData)
34+
qr.save(args.out + '/' + str(book['id']) + '.png')
35+
else:
36+
print('File or directory not exists!')
37+
38+
if __name__ == '__main__':
39+
sys.exit(main() or 0)

‎demos/smart_library_demo/library.json ‎demos/python_demos/smart_library_demo/library.json

File renamed without changes.

‎demos/python_demos/smart_library_demo/models.lst

+5
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
# This file can be used with the --list option of the model downloader.
2+
face-detection-retail-????
3+
face-reidentification-retail-????
4+
landmarks-regression-retail-????
5+

‎demos/python_demos/smart_library_demo/requirements.txt

+2
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,2 @@
1+
pyzbar>=0.1.8
2+
qrcode>=6.1

0 commit comments

Comments
 (0)