OpenVINO on Raspberry Pi 4 with Movidius Neural Compute Stick II

Posted on Edited on In ,

Today is Easter Sunday, Prime minister of United Kindom Boris Johnson recovered from COVID-19. Canada has been suffereing COVID-19 for a month already. I'm re-writing this blog FIRSTLY written in September 2019, and UPDATED ONCE in December 2019.

Merry Christmas and happy new year everybody. I’ve been back to Vancouver for several days. These days, I’m updating this blog FIRSTLY written in September 2019. 2020 is coming, and we’re getting 1 year older. A kind of sad hum?

😝

Okay… No matter what, let’s enjoy the song first: WE ARE YOUNG. Today, I joined Free Software Foundation and start my journey of supporting Open Source Software BY CASH. For me, it’s not about poverity or richness. It’s ALL about FAITH.

To write something about Raspberry Pi is to say GOOD BYE to my Raspberry Pi 3B, and WELCOME Raspberry Pi 4 at the same time. Our target today is to build an AI edge computing end as the following Youtube video:

1. About Raspberry Pi 4

1.1 Raspberry Pi 4 vs. Raspberry Pi 3B+

Before we start, let’s carry out a simple comparison between Raspberry Pi 4 and Raspberry Pi 3B+.

1.2 Raspbian Installation

1
2
3
4
5
➜  raspbian sudo dd bs=4M if=2020-02-13-raspbian-buster.img of=/dev/mmcblk0 conv=fsync 
[sudo] password for longervision: 
903+0 records in
903+0 records out
3787456512 bytes (3.8 GB, 3.5 GiB) copied, 198.861 s, 19.0 MB/s

1.3 BCM2711 is detected as BCM2835

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
➜  ~ cat /proc/cpuinfo
processor	: 0
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 108.00
Features	: half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xd08
CPU revision	: 3

processor	: 1
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 108.00
Features	: half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xd08
CPU revision	: 3

processor	: 2
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 108.00
Features	: half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xd08
CPU revision	: 3

processor	: 3
model name	: ARMv7 Processor rev 3 (v7l)
BogoMIPS	: 108.00
Features	: half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm crc32 
CPU implementer	: 0x41
CPU architecture: 7
CPU variant	: 0x0
CPU part	: 0xd08
CPU revision	: 3

Hardware	: BCM2835
Revision	: c03111
Serial		: 100000006c0c9b01
Model		: Raspberry Pi 4 Model B Rev 1.1

This issue seems to be a well-known bug. Raspberry Pi 4’s specification can be retrieved from The MagPi Magazine. More details about the development history of Raspberry Pi can be found on Wikipedia.

2. Movidius Neural Compute Stick on Raspberry Pi 4

Then, we just follow the following 2 blogs Run NCS Applications on Raspberry Pi and Adding AI to the Raspberry Pi with the Movidius Neural Compute Stick to test out Intel Movidius Neural Compute Stick 2:

Intel Movidius Neural Compute Stick 2 Intel Movidius Neural Compute Stick 1
Intel Movidius Neural Compute Stick 2 Intel Movidius Neural Compute Stick 1

Intel Movidius Neural Compute Stick 1 is NOT listed on Intel’s official website any more. But github support for Intel Movidius Neural Compute Stick 1 can be found at https://github.com/movidius/ncsdk.

2.1 NCSDK Installation

We FIRST need to have ncsdk installed. Yup, here, as described in Run NCS Applications on Raspberry Pi, we carry out the installation directly under folder ...../ncsdk/api/src.

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
➜  src make
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c mvnc_api.c -o obj-armv7l/mvnc_api.o
mvnc_api.c:457:12: warning: ‘deviceGetNumberOfDevices’ defined but not used [-Wunused-function]
 static int deviceGetNumberOfDevices()
            ^~~~~~~~~~~~~~~~~~~~~~~~
mvnc_api.c: In function ‘ncGraphCreate’:
mvnc_api.c:898:5: warning: ‘strncpy’ specified bound 28 equals destination size [-Wstringop-truncation]
     strncpy(g->name, name, NC_MAX_NAME_SIZE);
     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
mvnc_api.c: In function ‘ncFifoCreate’:
mvnc_api.c:2210:5: warning: ‘strncpy’ specified bound 28 equals destination size [-Wstringop-truncation]
     strncpy(handle->name, name, NC_MAX_NAME_SIZE);
     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c fp16.c -o obj-armv7l/fp16.o
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/usb_boot.c -o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/usb_boot.o
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/pcie_host.c -o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/pcie_host.o
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLink.c -o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLink.o
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLinkDispatcher.c -o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLinkDispatcher.o
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.c -o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.o
/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.c: In function ‘shellThreadWriter’:
/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.c:44:33: warning: pointer targets in passing argument 2 of ‘XLinkWriteData’ differ in signedness [-Wpointer-sign]
             XLinkWriteData(cId, str, bytes);
                                 ^~~
In file included from /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.c:18:
/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLink.h:54:14: note: expected ‘const uint8_t *’ {aka ‘const unsigned char *’} but argument is of type ‘char *’
 XLinkError_t XLinkWriteData(streamId_t streamId, const uint8_t* buffer, int size);
              ^~~~~~~~~~~~~~
/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.c: In function ‘shellThreadReader’:
/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.c:81:5: warning: implicit declaration of function ‘pthread_create’ [-Wimplicit-function-declaration]
     pthread_create(&shellWriter, NULL, shellThreadWriter, (void*) context);
     ^~~~~~~~~~~~~~
cc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc -I/home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/swCommon/include/ -I /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/shared/include/ -D__PC__ -DUSE_USB_VSC -DVERSION_NAME="\"`cat ./version.txt`\"" -DDEVICE_SHELL_ENABLED -DEXCLUDE_HIGHCLASS -O2 -Wall -pthread -fPIC -MMD -MP -I. -I../include -I/usr/include/libusb-1.0  -c /home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/XLinkPlatform.c -o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/XLinkPlatform.o
if [ ! -e ./version.txt ] ; then echo "missing version.txt file"; exit 1; fi;
cc -shared obj-armv7l/mvnc_api.o obj-armv7l/fp16.o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/usb_boot.o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/pcie_host.o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLink.o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/shared/XLinkDispatcher.o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLinkConsole/pc/XLinkConsole.o obj-armv7l//home/pi/Downloads/ncsdk-2.10.01.01/api/src/common/components/XLink/pc/XLinkPlatform.o -o obj-armv7l/libmvnc.so.0 -lpthread -lusb-1.0 -ldl
ln -fs obj-armv7l/libmvnc.so.0 libmvnc.so
ln -fs obj-armv7l/libmvnc.so.0 libmvnc.so.0
NCSDK FW successfully installed
➜  src sudo make install
NCSDK FW successfully installed
mkdir -p /usr/local/include/
mkdir -p /usr/local/include/mvnc2
mkdir -p /usr/local/lib/
cp obj-armv7l/libmvnc.so.0 /usr/local/lib/
ln -fs libmvnc.so.0 /usr/local/lib/libmvnc.so
cp ../include/mvnc.h /usr/local/include/mvnc2
ln -fs /usr/local/include/mvnc2/mvnc.h /usr/local/include/mvnc.h 
mkdir -p /usr/local/lib/mvnc
cp mvnc/MvNCAPI-*.mvcmd /usr/local/lib/mvnc/
mkdir -p /etc/udev/rules.d/
cp 97-usbboot.rules /etc/udev/rules.d/
mkdir -p /usr/local/lib/python3.7/dist-packages
mkdir -p /usr/local/lib/python3.7/dist-packages
cp -r ../python/mvnc /usr/local/lib/python3.7/dist-packages/
cp -r ../python/mvnc /usr/local/lib/python3.7/dist-packages/
udevadm control --reload-rules
udevadm trigger
ldconfig
➜  src

2.2 Test NCSDK Example Apps

2.2.1 For Movidius NCS 1

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
➜  hello_ncs_py lsusb
......
Bus 001 Device 004: ID 03e7:2150 Intel Myriad VPU [Movidius Neural Compute Stick]
......
➜  hello_ncs_py python hello_ncs.py
D: [         0] ncDeviceCreate:307	ncDeviceCreate index 0
D: [         0] ncDeviceCreate:307	ncDeviceCreate index 1
D: [         0] ncDeviceOpen:523	File path /usr/local/lib/mvnc/MvNCAPI-ma2450.mvcmd
I: [         0] ncDeviceOpen:529	ncDeviceOpen() XLinkBootRemote returned success 0
I: [         0] ncDeviceOpen:567	XLinkConnect done - link Id 0
D: [         0] ncDeviceOpen:581	done
I: [         0] ncDeviceOpen:583	Booted 1.1-ma2450 -> VSC
I: [         0] getDevAttributes:382	Device attributes
I: [         0] getDevAttributes:385	Device FW version: 2.a.2450.8a
I: [         0] getDevAttributes:387	mvTensorVersion 2.10
I: [         0] getDevAttributes:388	Maximum graphs: 10
I: [         0] getDevAttributes:389	Maximum fifos: 20
I: [         0] getDevAttributes:391	Maximum graph option class: 1
I: [         0] getDevAttributes:393	Maximum device option class: 1
I: [         0] getDevAttributes:394	Device memory capacity: 522047856
Hello NCS! Device opened normally.
I: [         0] ncDeviceClose:775	closing device
Goodbye NCS! Device closed normally.
NCS device working.

2.2.2 For Movidius NCS 2

1
2
3
4
5
6
7
8
9
10
➜  hello_ncs_py lsusb
......
Bus 001 Device 007: ID 03e7:2485 Intel Movidius MyriadX
......
➜  hello_ncs_py python hello_ncs.py 
D: [         0] ncDeviceCreate:307	ncDeviceCreate index 0
D: [         0] ncDeviceCreate:307	ncDeviceCreate index 1
D: [         0] ncDeviceOpen:523	File path /usr/local/lib/mvnc/MvNCAPI-ma2480.mvcmd
W: [         0] ncDeviceOpen:527	ncDeviceOpen() XLinkBootRemote returned error 3
Error - Could not open NCS device.

The above bug has ALREADY been expalined in main online resource:

All these hint that OpenVINO should be utilized instead of NCSDK2.

2.3 mvnc Python Package

1
2
3
4
5
6
7
8
➜  ~ python
Python 3.7.3 (default, Dec 20 2019, 18:57:59) 
[GCC 8.3.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import mvnc
>>> mvnc.__file__
'/usr/local/lib/python3.7/dist-packages/mvnc/__init__.py'
>>>

3. Transitioning from Intel Movidius Neural Compute SDK to Intel OpenVINO

By following Intel’s official documentation Transitioning from Intel® Movidius™ Neural Compute SDK to Intel® Distribution of OpenVINO™ toolkit, we are transitioning to OpenVINO, which supports both Intel NCS 2 and Intel NCS 1.

3.1 Install OpenVINO for Raspbian

For the installation details of OpenVINO, please refer to the following 2 documentations:

We now extract the MOST up-to-date l_openvino_toolkit_runtime_raspbian_p_2020.3.220.tgz under folder /opt/intel/openvino. Let’s take a brief look at this folder:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
➜  openvino pwd
/opt/intel/openvino
➜  openvino ll
total 28K
drwxr-xr-x 2 pi pi 4.0K Jan 27 08:06 bin
drwxr-xr-x 4 pi pi 4.0K Jan 27 08:06 deployment_tools
drwxr-xr-x 2 pi pi 4.0K Jan 27 08:08 documentation
lrwxrwxrwx 1 pi pi   33 Jan 27 08:04 inference_engine -> deployment_tools/inference_engine
drwxr-xr-x 2 pi pi 4.0K Jan 27 08:08 install_dependencies
drwxr-xr-x 5 pi pi 4.0K Jan 27 08:08 licensing
drwxr-xr-x 8 pi pi 4.0K Jan 27 08:09 opencv
drwxr-xr-x 6 pi pi 4.0K Jan 27 08:09 python
➜  openvino ls inference_engine 
external  include  lib  samples  share  version.txt

Clearly, by comparing with OpenVINO™ Toolkit - Deep Learning Deployment Toolkit repository, we know that the open source version of deployment_tools contains some more content than the trimmed version for Raspbian. We’ll use model-optimizer for sure. Therefore, we checked out dldt, and put it under folder /opt/intel.

1
2
3
4
5
6
7
➜  intel pwd
/opt/intel
➜  intel ll
total 12K
drwxr-xr-x 8 pi   pi   4.0K Apr  7 06:37 dldt
drwxr-xr-x 9 pi   pi   4.0K Apr  7 14:04 l_openvino_toolkit_runtime_raspbian_p_2020.1.023
lrwxrwxrwx 1 root root   48 Apr  7 06:29 openvino -> l_openvino_toolkit_runtime_raspbian_p_2020.1.023

3.2 Build OpenVINO Samples

Before start building OpenVINO samples, please have OpenCV built from source and installed. You can of course directly run the command build_samples.sh to build ALL samples. However, I personally would like you to build and install OpenCV from source FIRST.

Note: Be sure to enable -DCMAKE_CXX_FLAGS=’-march=armv7-a’ while building dldt/samples, which is exactly the same as the source under l_openvino_toolkit_runtime_raspbian_p_2020.1.023/inference_engine/samples/cpp. In fact, to build l_openvino_toolkit_runtime_raspbian_p_2020.1.023/inference_engine/samples/c, -DCMAKE_CXX_FLAGS=’-march=armv7-a’ also needs to be enabled.

After having successfully built C/C++ samples, let's enter folder /opt/intel/openvino/inference_engine/samples.

3.3 Device Query

3.2.1 C

There is NO such an exe file hello_query_device_c.

3.2.2 C++

For NCS 2

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
➜  samples $ ./cpp/build/armv7l/Release/hello_query_device
Available devices: 
	Device: MYRIAD
	Metrics: 
		DEVICE_THERMAL : UNSUPPORTED TYPE
		RANGE_FOR_ASYNC_INFER_REQUESTS : { 3, 6, 1 }
		SUPPORTED_CONFIG_KEYS : [ CONFIG_FILE PERF_COUNT EXCLUSIVE_ASYNC_REQUESTS DEVICE_ID VPU_MYRIAD_PLATFORM VPU_IGNORE_IR_STATISTIC VPU_CUSTOM_LAYERS VPU_MYRIAD_FORCE_RESET VPU_PRINT_RECEIVE_TENSOR_TIME LOG_LEVEL VPU_HW_STAGES_OPTIMIZATION ]
		SUPPORTED_METRICS : [ DEVICE_THERMAL RANGE_FOR_ASYNC_INFER_REQUESTS SUPPORTED_CONFIG_KEYS SUPPORTED_METRICS OPTIMIZATION_CAPABILITIES FULL_DEVICE_NAME AVAILABLE_DEVICES ]
		OPTIMIZATION_CAPABILITIES : [ FP16 ]
		FULL_DEVICE_NAME : Intel Movidius Myriad X VPU
	Default values for device configuration keys: 
		CONFIG_FILE : ""
		PERF_COUNT : OFF
		EXCLUSIVE_ASYNC_REQUESTS : OFF
		DEVICE_ID : ""
		VPU_MYRIAD_PLATFORM : ""
		VPU_IGNORE_IR_STATISTIC : OFF
		VPU_CUSTOM_LAYERS : ""
		VPU_MYRIAD_FORCE_RESET : OFF
		VPU_PRINT_RECEIVE_TENSOR_TIME : OFF
		LOG_LEVEL : LOG_NONE
		VPU_HW_STAGES_OPTIMIZATION : ON

For NCS 1

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
➜  samples ./cpp/build/armv7l/Release/hello_query_device
Available devices: 
	Device: MYRIAD
	Metrics: 
		DEVICE_THERMAL : UNSUPPORTED TYPE
		RANGE_FOR_ASYNC_INFER_REQUESTS : { 3, 6, 1 }
		SUPPORTED_CONFIG_KEYS : [ CONFIG_FILE PERF_COUNT EXCLUSIVE_ASYNC_REQUESTS DEVICE_ID VPU_MYRIAD_PLATFORM VPU_IGNORE_IR_STATISTIC VPU_CUSTOM_LAYERS VPU_MYRIAD_FORCE_RESET VPU_PRINT_RECEIVE_TENSOR_TIME LOG_LEVEL VPU_HW_STAGES_OPTIMIZATION ]
		SUPPORTED_METRICS : [ DEVICE_THERMAL RANGE_FOR_ASYNC_INFER_REQUESTS SUPPORTED_CONFIG_KEYS SUPPORTED_METRICS OPTIMIZATION_CAPABILITIES FULL_DEVICE_NAME AVAILABLE_DEVICES ]
		OPTIMIZATION_CAPABILITIES : [ FP16 ]
		FULL_DEVICE_NAME : Intel Movidius Myriad 2 VPU
	Default values for device configuration keys: 
		CONFIG_FILE : ""
		PERF_COUNT : OFF
		EXCLUSIVE_ASYNC_REQUESTS : OFF
		DEVICE_ID : ""
		VPU_MYRIAD_PLATFORM : ""
		VPU_IGNORE_IR_STATISTIC : OFF
		VPU_CUSTOM_LAYERS : ""
		VPU_MYRIAD_FORCE_RESET : OFF
		VPU_PRINT_RECEIVE_TENSOR_TIME : OFF
		LOG_LEVEL : LOG_NONE
		VPU_HW_STAGES_OPTIMIZATION : ON

3.2.3 Python

For NCS 2

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
➜  samples python ./python/hello_query_device/hello_query_device.py
Available devices:
	Device: MYRIAD
	Metrics:
		DEVICE_THERMAL: UNSUPPORTED TYPE
		RANGE_FOR_ASYNC_INFER_REQUESTS: 3, 6, 1
		SUPPORTED_CONFIG_KEYS: CONFIG_FILE, PERF_COUNT, EXCLUSIVE_ASYNC_REQUESTS, DEVICE_ID, VPU_MYRIAD_PLATFORM, VPU_IGNORE_IR_STATISTIC, VPU_CUSTOM_LAYERS, VPU_MYRIAD_FORCE_RESET, VPU_PRINT_RECEIVE_TENSOR_TIME, LOG_LEVEL, VPU_HW_STAGES_OPTIMIZATION
		SUPPORTED_METRICS: DEVICE_THERMAL, RANGE_FOR_ASYNC_INFER_REQUESTS, SUPPORTED_CONFIG_KEYS, SUPPORTED_METRICS, OPTIMIZATION_CAPABILITIES, FULL_DEVICE_NAME, AVAILABLE_DEVICES
		OPTIMIZATION_CAPABILITIES: FP16
		FULL_DEVICE_NAME: Intel Movidius Myriad X VPU
		AVAILABLE_DEVICES: 1.1.1-ma2480

	Default values for device configuration keys:
		CONFIG_FILE: 
		PERF_COUNT: OFF
		EXCLUSIVE_ASYNC_REQUESTS: OFF
		DEVICE_ID: 
		VPU_MYRIAD_PLATFORM: 
		VPU_IGNORE_IR_STATISTIC: OFF
		VPU_CUSTOM_LAYERS: 
		VPU_MYRIAD_FORCE_RESET: OFF
		VPU_PRINT_RECEIVE_TENSOR_TIME: OFF
		LOG_LEVEL: LOG_NONE
		VPU_HW_STAGES_OPTIMIZATION: ON

For NCS 1

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
➜  samples python ./python/hello_query_device/hello_query_device.py
Available devices:
	Device: MYRIAD
	Metrics:
		DEVICE_THERMAL: UNSUPPORTED TYPE
		RANGE_FOR_ASYNC_INFER_REQUESTS: 3, 6, 1
		SUPPORTED_CONFIG_KEYS: CONFIG_FILE, PERF_COUNT, EXCLUSIVE_ASYNC_REQUESTS, DEVICE_ID, VPU_MYRIAD_PLATFORM, VPU_IGNORE_IR_STATISTIC, VPU_CUSTOM_LAYERS, VPU_MYRIAD_FORCE_RESET, VPU_PRINT_RECEIVE_TENSOR_TIME, LOG_LEVEL, VPU_HW_STAGES_OPTIMIZATION
		SUPPORTED_METRICS: DEVICE_THERMAL, RANGE_FOR_ASYNC_INFER_REQUESTS, SUPPORTED_CONFIG_KEYS, SUPPORTED_METRICS, OPTIMIZATION_CAPABILITIES, FULL_DEVICE_NAME, AVAILABLE_DEVICES
		OPTIMIZATION_CAPABILITIES: FP16
		FULL_DEVICE_NAME: Intel Movidius Myriad 2 VPU
		AVAILABLE_DEVICES: 1.1.1-ma2450

	Default values for device configuration keys:
		CONFIG_FILE: 
		PERF_COUNT: OFF
		EXCLUSIVE_ASYNC_REQUESTS: OFF
		DEVICE_ID: 
		VPU_MYRIAD_PLATFORM: 
		VPU_IGNORE_IR_STATISTIC: OFF
		VPU_CUSTOM_LAYERS: 
		VPU_MYRIAD_FORCE_RESET: OFF
		VPU_PRINT_RECEIVE_TENSOR_TIME: OFF
		LOG_LEVEL: LOG_NONE
		VPU_HW_STAGES_OPTIMIZATION: ON

3.3 Object Detection

Please refer to Device-specific Plugin Libraries for ALL possible device types.

We then download two model files as given in the blog Install OpenVINO™ toolkit for Raspbian* OS.

Please note that:

Afterwards, start running Object Detection for 2 images: me.jpg and parents.jpg:

3.3.1 C

For NCS 2

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
➜  samples ./c/build/armv7l/Release/object_detection_sample_ssd_c -m face-detection-adas-0001.xml -d MYRIAD -i parents.jpg
[ INFO ] InferenceEngine: 
2.1.2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     parents.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
         MYRIAD
         myriadPlugin version ......... 2.1
         Build ......... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image is resized from (1280, 960) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ INFO ] Start inference
[ INFO ] Processing output blobs
[0, 1] element, prob = 1.000000    (826, 366)-(1026, 644) batch id : 0 WILL BE PRINTED!
[1, 1] element, prob = 0.996582    (539, 173)-(693, 429) batch id : 0 WILL BE PRINTED!
[2, 1] element, prob = 0.539062    (1094, 47)-(1135, 96) batch id : 0 WILL BE PRINTED!
[3, 1] element, prob = 0.212402    (848, 22)-(886, 70) batch id : 0
[4, 1] element, prob = 0.052734    (7, 783)-(151, 957) batch id : 0
[5, 1] element, prob = 0.039551    (1033, 78)-(1070, 122) batch id : 0
[6, 1] element, prob = 0.034668    (1123, 75)-(1158, 122) batch id : 0
[7, 1] element, prob = 0.032227    (1091, 109)-(1130, 171) batch id : 0
[8, 1] element, prob = 0.032227    (1086, 14)-(1138, 85) batch id : 0
[9, 1] element, prob = 0.030273    (1109, 43)-(1150, 93) batch id : 0
[10, 1] element, prob = 0.029297    (1151, 83)-(1190, 131) batch id : 0
[11, 1] element, prob = 0.027344    (1030, 2)-(1073, 43) batch id : 0
[12, 1] element, prob = 0.027344    (1091, 6)-(1133, 55) batch id : 0
[13, 1] element, prob = 0.027344    (1064, 73)-(1098, 121) batch id : 0
[14, 1] element, prob = 0.027344    (1058, 113)-(1096, 168) batch id : 0
[15, 1] element, prob = 0.027344    (1143, 74)-(1211, 165) batch id : 0
[16, 1] element, prob = 0.026855    (876, 73)-(915, 118) batch id : 0
[17, 1] element, prob = 0.026855    (1047, -3)-(1110, 59) batch id : 0
[18, 1] element, prob = 0.026855    (849, 13)-(1030, 236) batch id : 0
[19, 1] element, prob = 0.025879    (941, 22)-(981, 74) batch id : 0
[20, 1] element, prob = 0.025879    (1030, 35)-(1074, 89) batch id : 0
[21, 1] element, prob = 0.025879    (845, 112)-(886, 167) batch id : 0
[22, 1] element, prob = 0.025879    (1036, 111)-(1076, 171) batch id : 0
[23, 1] element, prob = 0.025879    (1113, -1)-(1175, 59) batch id : 0
[24, 1] element, prob = 0.025879    (1002, 40)-(1093, 151) batch id : 0
[25, 1] element, prob = 0.024902    (1059, 0)-(1098, 43) batch id : 0
[26, 1] element, prob = 0.024902    (876, 29)-(918, 79) batch id : 0
[27, 1] element, prob = 0.024902    (1002, 69)-(1048, 123) batch id : 0
[28, 1] element, prob = 0.024902    (1133, 104)-(1170, 163) batch id : 0
[29, 1] element, prob = 0.024902    (1056, 145)-(1094, 208) batch id : 0
[30, 1] element, prob = 0.024902    (861, 2)-(918, 63) batch id : 0
[31, 1] element, prob = 0.023926    (966, 1)-(1018, 45) batch id : 0
[32, 1] element, prob = 0.023926    (1098, 63)-(1131, 109) batch id : 0
[33, 1] element, prob = 0.023926    (1010, 84)-(1092, 204) batch id : 0
[34, 1] element, prob = 0.023926    (1022, 139)-(1076, 224) batch id : 0
[35, 1] element, prob = 0.023926    (943, 34)-(1073, 208) batch id : 0
[36, 1] element, prob = 0.022949    (811, 3)-(851, 48) batch id : 0
[37, 1] element, prob = 0.022949    (1154, 27)-(1198, 79) batch id : 0
[38, 1] element, prob = 0.022949    (1000, 155)-(1046, 226) batch id : 0
[39, 1] element, prob = 0.022949    (1086, 145)-(1125, 208) batch id : 0
[40, 1] element, prob = 0.022949    (958, 20)-(1030, 84) batch id : 0
[41, 1] element, prob = 0.022949    (934, 65)-(986, 155) batch id : 0
[42, 1] element, prob = 0.022949    (1045, 99)-(1106, 178) batch id : 0
[43, 1] element, prob = 0.022949    (983, -9)-(1141, 105) batch id : 0
[44, 1] element, prob = 0.022949    (1065, -16)-(1192, 124) batch id : 0
[45, 1] element, prob = 0.022949    (922, 81)-(1087, 304) batch id : 0
[46, 1] element, prob = 0.021973    (796, 3)-(834, 47) batch id : 0
[47, 1] element, prob = 0.021973    (911, 4)-(953, 55) batch id : 0
[48, 1] element, prob = 0.021973    (797, 29)-(835, 79) batch id : 0
[49, 1] element, prob = 0.021973    (821, 34)-(855, 81) batch id : 0
[50, 1] element, prob = 0.021973    (910, 28)-(945, 78) batch id : 0
[51, 1] element, prob = 0.021973    (755, 77)-(790, 125) batch id : 0
[52, 1] element, prob = 0.021973    (969, 80)-(1008, 128) batch id : 0
[53, 1] element, prob = 0.021973    (938, 108)-(980, 173) batch id : 0
[54, 1] element, prob = 0.021973    (1123, 186)-(1161, 252) batch id : 0
[55, 1] element, prob = 0.021973    (985, 28)-(1056, 105) batch id : 0
[56, 1] element, prob = 0.021973    (845, 56)-(898, 135) batch id : 0
[57, 1] element, prob = 0.021973    (1023, 62)-(1080, 131) batch id : 0
[58, 1] element, prob = 0.021973    (1041, 40)-(1125, 143) batch id : 0
[59, 1] element, prob = 0.021973    (1035, 116)-(1110, 242) batch id : 0
[60, 1] element, prob = 0.021973    (990, 166)-(1054, 269) batch id : 0
[61, 1] element, prob = 0.021973    (1078, 169)-(1130, 261) batch id : 0
[62, 1] element, prob = 0.021973    (971, 13)-(1140, 235) batch id : 0
[63, 1] element, prob = 0.021973    (980, 125)-(1108, 286) batch id : 0
[64, 1] element, prob = 0.020996    (940, 5)-(987, 55) batch id : 0
[65, 1] element, prob = 0.020996    (991, 0)-(1043, 41) batch id : 0
[66, 1] element, prob = 0.020996    (1126, 0)-(1165, 42) batch id : 0
[67, 1] element, prob = 0.020996    (998, 41)-(1039, 86) batch id : 0
[68, 1] element, prob = 0.020996    (1068, 40)-(1103, 86) batch id : 0
[69, 1] element, prob = 0.020996    (821, 77)-(856, 123) batch id : 0
[70, 1] element, prob = 0.020996    (913, 80)-(949, 128) batch id : 0
[71, 1] element, prob = 0.020996    (821, 112)-(861, 164) batch id : 0
[72, 1] element, prob = 0.020996    (878, 113)-(919, 175) batch id : 0
[73, 1] element, prob = 0.020996    (910, 110)-(950, 170) batch id : 0
[74, 1] element, prob = 0.020996    (1149, 111)-(1189, 163) batch id : 0
[75, 1] element, prob = 0.020996    (870, 149)-(918, 215) batch id : 0
[76, 1] element, prob = 0.020996    (1033, 152)-(1070, 209) batch id : 0
[77, 1] element, prob = 0.020996    (1128, 141)-(1165, 204) batch id : 0
[78, 1] element, prob = 0.020996    (1147, 135)-(1188, 199) batch id : 0
[79, 1] element, prob = 0.020996    (1064, 188)-(1099, 246) batch id : 0
[80, 1] element, prob = 0.020996    (1238, 903)-(1279, 963) batch id : 0
[81, 1] element, prob = 0.020996    (1146, 19)-(1208, 109) batch id : 0
[82, 1] element, prob = 0.020996    (1054, 58)-(1110, 126) batch id : 0
[83, 1] element, prob = 0.020996    (1111, 99)-(1190, 228) batch id : 0
[84, 1] element, prob = 0.020996    (1052, 161)-(1105, 265) batch id : 0
[85, 1] element, prob = 0.020996    (1060, 145)-(1139, 284) batch id : 0
[86, 1] element, prob = 0.020996    (-16, 860)-(79, 985) batch id : 0
[87, 1] element, prob = 0.020996    (1174, -17)-(1281, 136) batch id : 0
[88, 1] element, prob = 0.020996    (1073, 32)-(1198, 190) batch id : 0
[89, 1] element, prob = 0.020996    (950, 188)-(1072, 366) batch id : 0
[90, 1] element, prob = 0.020020    (943, 81)-(978, 130) batch id : 0
[91, 1] element, prob = 0.020020    (968, 104)-(1008, 159) batch id : 0
[92, 1] element, prob = 0.020020    (930, 148)-(973, 211) batch id : 0
[93, 1] element, prob = 0.020020    (1092, 191)-(1128, 246) batch id : 0
[94, 1] element, prob = 0.020020    (1124, 225)-(1161, 291) batch id : 0
[95, 1] element, prob = 0.020020    (1007, -2)-(1083, 55) batch id : 0
[96, 1] element, prob = 0.020020    (890, 60)-(960, 138) batch id : 0
[97, 1] element, prob = 0.020020    (962, 65)-(1023, 137) batch id : 0
[98, 1] element, prob = 0.020020    (1101, 145)-(1183, 284) batch id : 0
[99, 1] element, prob = 0.020020    (866, -10)-(1029, 107) batch id : 0
[100, 1] element, prob = 0.020020    (1005, 179)-(1111, 360) batch id : 0
[101, 1] element, prob = 0.020020    (1019, 342)-(1283, 785) batch id : 0
[102, 1] element, prob = 0.019043    (758, 2)-(795, 45) batch id : 0
[103, 1] element, prob = 0.019043    (756, 38)-(792, 89) batch id : 0
[104, 1] element, prob = 0.019043    (968, 36)-(1011, 82) batch id : 0
[105, 1] element, prob = 0.019043    (729, 77)-(763, 127) batch id : 0
[106, 1] element, prob = 0.019043    (1003, 109)-(1044, 169) batch id : 0
[107, 1] element, prob = 0.019043    (1030, 190)-(1066, 242) batch id : 0
[108, 1] element, prob = 0.019043    (887, 83)-(971, 209) batch id : 0
[109, 1] element, prob = 0.019043    (995, 99)-(1055, 192) batch id : 0
[110, 1] element, prob = 0.019043    (902, 135)-(963, 221) batch id : 0
[ INFO ] Image out_0.bmp created!
[ INFO ] Execution successful
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
➜  samples ./c/build/armv7l/Release/object_detection_sample_ssd_c -m face-detection-adas-0001.xml -d MYRIAD -i me.jpg     
[ INFO ] InferenceEngine: 
2.1.2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     me.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
         MYRIAD
         myriadPlugin version ......... 2.1
         Build ......... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image is resized from (924, 1280) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ INFO ] Start inference
[ INFO ] Processing output blobs
[0, 1] element, prob = 0.870117    (421, 202)-(745, 608) batch id : 0 WILL BE PRINTED!
[1, 1] element, prob = 0.026855    (-55, 556)-(234, 1100) batch id : 0
[2, 1] element, prob = 0.021973    (20, 925)-(239, 1340) batch id : 0
[3, 1] element, prob = 0.020020    (-18, -144)-(179, 581) batch id : 0
[4, 1] element, prob = 0.018066    (636, 315)-(757, 471) batch id : 0
[5, 1] element, prob = 0.017090    (-27, 672)-(137, 1217) batch id : 0
[6, 1] element, prob = 0.017090    (74, 976)-(364, 1322) batch id : 0
[7, 1] element, prob = 0.016113    (291, 958)-(349, 1122) batch id : 0
[8, 1] element, prob = 0.016113    (329, 963)-(391, 1126) batch id : 0
[9, 1] element, prob = 0.016113    (203, 928)-(423, 1346) batch id : 0
[10, 1] element, prob = 0.015625    (694, 1034)-(732, 1140) batch id : 0
[11, 1] element, prob = 0.015625    (832, 1121)-(937, 1295) batch id : 0
[12, 1] element, prob = 0.015625    (219, 39)-(388, 627) batch id : 0
[13, 1] element, prob = 0.015625    (76, 785)-(357, 1160) batch id : 0
[14, 1] element, prob = 0.014648    (355, 616)-(475, 778) batch id : 0
[15, 1] element, prob = 0.014648    (240, 779)-(331, 1048) batch id : 0
[16, 1] element, prob = 0.013672    (0, 893)-(24, 984) batch id : 0
[17, 1] element, prob = 0.013672    (658, 306)-(752, 405) batch id : 0
[18, 1] element, prob = 0.013672    (204, 1198)-(267, 1275) batch id : 0
[19, 1] element, prob = 0.013672    (322, 611)-(430, 794) batch id : 0
[20, 1] element, prob = 0.013672    (171, 244)-(423, 798) batch id : 0
[21, 1] element, prob = 0.013672    (252, 798)-(539, 1177) batch id : 0
[22, 1] element, prob = 0.013672    (133, 875)-(301, 1420) batch id : 0
[23, 1] element, prob = 0.013672    (443, 202)-(802, 1022) batch id : 0
[24, 1] element, prob = 0.012695    (718, 466)-(757, 540) batch id : 0
[25, 1] element, prob = 0.012695    (166, 527)-(206, 594) batch id : 0
[26, 1] element, prob = 0.012695    (694, 511)-(731, 590) batch id : 0
[27, 1] element, prob = 0.012695    (705, 441)-(771, 555) batch id : 0
[28, 1] element, prob = 0.012695    (272, 962)-(322, 1117) batch id : 0
[29, 1] element, prob = 0.012695    (268, 486)-(376, 688) batch id : 0
[30, 1] element, prob = 0.012695    (323, 489)-(422, 690) batch id : 0
[31, 1] element, prob = 0.012695    (277, 607)-(384, 796) batch id : 0
[32, 1] element, prob = 0.012695    (233, 709)-(342, 924) batch id : 0
[33, 1] element, prob = 0.012695    (256, 672)-(406, 957) batch id : 0
[34, 1] element, prob = 0.012695    (252, 1043)-(431, 1214) batch id : 0
[35, 1] element, prob = 0.011719    (695, 466)-(733, 539) batch id : 0
[36, 1] element, prob = 0.011719    (673, 514)-(709, 587) batch id : 0
[37, 1] element, prob = 0.011719    (370, 1005)-(399, 1078) batch id : 0
[38, 1] element, prob = 0.011719    (670, 256)-(753, 346) batch id : 0
[39, 1] element, prob = 0.011719    (151, 500)-(221, 610) batch id : 0
[40, 1] element, prob = 0.011719    (447, 489)-(511, 608) batch id : 0
[41, 1] element, prob = 0.011719    (126, 625)-(200, 719) batch id : 0
[42, 1] element, prob = 0.011719    (299, 626)-(387, 709) batch id : 0
[43, 1] element, prob = 0.011719    (164, 673)-(251, 747) batch id : 0
[44, 1] element, prob = 0.011719    (398, 850)-(453, 1008) batch id : 0
[45, 1] element, prob = 0.011719    (425, 851)-(476, 1008) batch id : 0
[46, 1] element, prob = 0.011719    (310, 915)-(371, 1077) batch id : 0
[47, 1] element, prob = 0.011719    (357, 925)-(413, 1073) batch id : 0
[48, 1] element, prob = 0.011719    (378, 923)-(435, 1074) batch id : 0
[49, 1] element, prob = 0.011719    (468, 906)-(521, 1070) batch id : 0
[50, 1] element, prob = 0.011719    (284, 1016)-(357, 1188) batch id : 0
[51, 1] element, prob = 0.011719    (203, -95)-(283, 182) batch id : 0
[52, 1] element, prob = 0.011719    (269, 378)-(374, 595) batch id : 0
[53, 1] element, prob = 0.011719    (-22, 575)-(71, 831) batch id : 0
[54, 1] element, prob = 0.011719    (318, 698)-(433, 911) batch id : 0
[55, 1] element, prob = 0.011719    (221, 716)-(395, 1258) batch id : 0
[56, 1] element, prob = 0.011719    (859, 771)-(987, 1518) batch id : 0
[57, 1] element, prob = 0.010742    (145, 466)-(185, 527) batch id : 0
[58, 1] element, prob = 0.010742    (166, 463)-(204, 521) batch id : 0
[59, 1] element, prob = 0.010742    (185, 477)-(224, 537) batch id : 0
[60, 1] element, prob = 0.010742    (214, 477)-(250, 540) batch id : 0
[61, 1] element, prob = 0.010742    (458, 478)-(488, 536) batch id : 0
[62, 1] element, prob = 0.010742    (151, 537)-(177, 594) batch id : 0
[63, 1] element, prob = 0.010742    (460, 522)-(488, 590) batch id : 0
[64, 1] element, prob = 0.010742    (481, 523)-(509, 588) batch id : 0
[65, 1] element, prob = 0.010742    (716, 512)-(755, 592) batch id : 0
[66, 1] element, prob = 0.010742    (129, 581)-(158, 641) batch id : 0
[67, 1] element, prob = 0.010742    (150, 579)-(179, 635) batch id : 0
[68, 1] element, prob = 0.010742    (171, 583)-(201, 641) batch id : 0
[69, 1] element, prob = 0.010742    (457, 564)-(491, 658) batch id : 0
[70, 1] element, prob = 0.010742    (481, 581)-(510, 647) batch id : 0
[71, 1] element, prob = 0.010742    (652, 565)-(689, 650) batch id : 0
[72, 1] element, prob = 0.010742    (674, 571)-(710, 650) batch id : 0
[73, 1] element, prob = 0.010742    (694, 571)-(734, 649) batch id : 0
[74, 1] element, prob = 0.010742    (143, 640)-(188, 705) batch id : 0
[75, 1] element, prob = 0.010742    (191, 743)-(224, 805) batch id : 0
[76, 1] element, prob = 0.010742    (437, 843)-(465, 920) batch id : 0
[77, 1] element, prob = 0.010742    (-1, 943)-(22, 1037) batch id : 0
[78, 1] element, prob = 0.010742    (414, 1007)-(443, 1080) batch id : 0
[79, 1] element, prob = 0.010742    (896, 1206)-(924, 1279) batch id : 0
[80, 1] element, prob = 0.010742    (678, 200)-(743, 294) batch id : 0
[81, 1] element, prob = 0.010742    (70, 374)-(190, 512) batch id : 0
[82, 1] element, prob = 0.010742    (422, 429)-(485, 563) batch id : 0
[83, 1] element, prob = 0.010742    (106, 477)-(213, 629) batch id : 0
[84, 1] element, prob = 0.010742    (428, 477)-(479, 615) batch id : 0
[85, 1] element, prob = 0.010742    (648, 498)-(727, 605) batch id : 0
[86, 1] element, prob = 0.010742    (703, 490)-(769, 614) batch id : 0
[87, 1] element, prob = 0.010742    (16, 531)-(150, 712) batch id : 0
[88, 1] element, prob = 0.010742    (87, 546)-(150, 685) batch id : 0
[89, 1] element, prob = 0.010742    (127, 552)-(197, 653) batch id : 0
[90, 1] element, prob = 0.010742    (425, 557)-(481, 671) batch id : 0
[91, 1] element, prob = 0.010742    (437, 509)-(516, 700) batch id : 0
[92, 1] element, prob = 0.010742    (467, 549)-(524, 678) batch id : 0
[93, 1] element, prob = 0.010742    (633, 551)-(704, 664) batch id : 0
[94, 1] element, prob = 0.010742    (416, 611)-(487, 735) batch id : 0
[95, 1] element, prob = 0.010742    (466, 602)-(530, 734) batch id : 0
[96, 1] element, prob = 0.010742    (124, 673)-(198, 759) batch id : 0
[97, 1] element, prob = 0.010742    (156, 705)-(260, 849) batch id : 0
[98, 1] element, prob = 0.010742    (54, 828)-(137, 933) batch id : 0
[99, 1] element, prob = 0.010742    (451, 812)-(498, 960) batch id : 0
[100, 1] element, prob = 0.010742    (470, 843)-(520, 1004) batch id : 0
[101, 1] element, prob = 0.010742    (447, 897)-(500, 1066) batch id : 0
[102, 1] element, prob = 0.010742    (641, 966)-(706, 1122) batch id : 0
[103, 1] element, prob = 0.010742    (164, 85)-(233, 410) batch id : 0
[104, 1] element, prob = 0.010742    (188, 497)-(307, 688) batch id : 0
[105, 1] element, prob = 0.010742    (205, 445)-(357, 715) batch id : 0
[106, 1] element, prob = 0.010742    (230, 602)-(341, 791) batch id : 0
[107, 1] element, prob = 0.010742    (274, 785)-(390, 1021) batch id : 0
[108, 1] element, prob = 0.010742    (26, 709)-(209, 1239) batch id : 0
[109, 1] element, prob = 0.010742    (309, 868)-(485, 1410) batch id : 0
[110, 1] element, prob = 0.010742    (229, 981)-(575, 1306) batch id : 0
[111, 1] element, prob = 0.010742    (29, 141)-(518, 883) batch id : 0
[ INFO ] Image out_0.bmp created!
[ INFO ] Execution successful

For NCS 1

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
➜  samples pwd
/opt/intel/openvino/inference_engine/samples
➜  samples ./c/build/armv7l/Release/object_detection_sample_ssd_c -m face-detection-adas-0001.xml -d MYRIAD -i parents.jpg
[ INFO ] InferenceEngine: 
2.1.2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     parents.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
         MYRIAD
         myriadPlugin version ......... 2.1
         Build ......... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image is resized from (1280, 960) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ INFO ] Start inference
[ INFO ] Processing output blobs
[0, 1] element, prob = 1.000000    (826, 366)-(1026, 644) batch id : 0 WILL BE PRINTED!
[1, 1] element, prob = 0.996582    (539, 173)-(693, 429) batch id : 0 WILL BE PRINTED!
[2, 1] element, prob = 0.552734    (1094, 47)-(1135, 95) batch id : 0 WILL BE PRINTED!
[3, 1] element, prob = 0.221680    (848, 22)-(886, 70) batch id : 0
[4, 1] element, prob = 0.053711    (8, 784)-(151, 956) batch id : 0
[5, 1] element, prob = 0.039551    (1033, 78)-(1070, 122) batch id : 0
[6, 1] element, prob = 0.034668    (1123, 75)-(1158, 122) batch id : 0
[7, 1] element, prob = 0.032227    (1086, 14)-(1138, 85) batch id : 0
[8, 1] element, prob = 0.031250    (1091, 110)-(1130, 171) batch id : 0
[9, 1] element, prob = 0.030273    (1108, 43)-(1150, 93) batch id : 0
[10, 1] element, prob = 0.029297    (1151, 83)-(1190, 131) batch id : 0
[11, 1] element, prob = 0.027344    (1091, 6)-(1133, 55) batch id : 0
[12, 1] element, prob = 0.027344    (1064, 73)-(1098, 121) batch id : 0
[13, 1] element, prob = 0.026855    (1030, 2)-(1073, 42) batch id : 0
[14, 1] element, prob = 0.026855    (1058, 113)-(1096, 168) batch id : 0
[15, 1] element, prob = 0.026855    (1047, -3)-(1110, 59) batch id : 0
[16, 1] element, prob = 0.026855    (1142, 75)-(1211, 165) batch id : 0
[17, 1] element, prob = 0.026855    (849, 14)-(1030, 236) batch id : 0
[18, 1] element, prob = 0.025879    (941, 22)-(981, 74) batch id : 0
[19, 1] element, prob = 0.025879    (1030, 36)-(1074, 89) batch id : 0
[20, 1] element, prob = 0.025879    (876, 73)-(915, 118) batch id : 0
[21, 1] element, prob = 0.025879    (845, 112)-(886, 167) batch id : 0
[22, 1] element, prob = 0.025879    (1036, 110)-(1076, 171) batch id : 0
[23, 1] element, prob = 0.025879    (1113, -1)-(1175, 59) batch id : 0
[24, 1] element, prob = 0.025879    (1003, 41)-(1094, 151) batch id : 0
[25, 1] element, prob = 0.024902    (1059, 0)-(1098, 43) batch id : 0
[26, 1] element, prob = 0.024902    (1002, 69)-(1048, 123) batch id : 0
[27, 1] element, prob = 0.024902    (1133, 105)-(1170, 164) batch id : 0
[28, 1] element, prob = 0.023926    (966, 1)-(1018, 45) batch id : 0
[29, 1] element, prob = 0.023926    (876, 29)-(918, 79) batch id : 0
[30, 1] element, prob = 0.023926    (1056, 145)-(1094, 208) batch id : 0
[31, 1] element, prob = 0.023926    (861, 2)-(918, 63) batch id : 0
[32, 1] element, prob = 0.023926    (1010, 84)-(1092, 204) batch id : 0
[33, 1] element, prob = 0.023926    (1022, 139)-(1076, 224) batch id : 0
[34, 1] element, prob = 0.023926    (942, 35)-(1072, 208) batch id : 0
[35, 1] element, prob = 0.022949    (811, 3)-(851, 48) batch id : 0
[36, 1] element, prob = 0.022949    (1098, 63)-(1131, 109) batch id : 0
[37, 1] element, prob = 0.022949    (983, -9)-(1141, 105) batch id : 0
[38, 1] element, prob = 0.022949    (921, 81)-(1087, 304) batch id : 0
[39, 1] element, prob = 0.021973    (796, 3)-(835, 47) batch id : 0
[40, 1] element, prob = 0.021973    (911, 4)-(953, 55) batch id : 0
[41, 1] element, prob = 0.021973    (797, 29)-(835, 79) batch id : 0
[42, 1] element, prob = 0.021973    (1154, 27)-(1198, 79) batch id : 0
[43, 1] element, prob = 0.021973    (969, 80)-(1008, 128) batch id : 0
[44, 1] element, prob = 0.021973    (1000, 155)-(1046, 226) batch id : 0
[45, 1] element, prob = 0.021973    (1086, 145)-(1125, 208) batch id : 0
[46, 1] element, prob = 0.021973    (1123, 186)-(1161, 252) batch id : 0
[47, 1] element, prob = 0.021973    (958, 20)-(1030, 84) batch id : 0
[48, 1] element, prob = 0.021973    (985, 28)-(1056, 105) batch id : 0
[49, 1] element, prob = 0.021973    (845, 56)-(898, 134) batch id : 0
[50, 1] element, prob = 0.021973    (934, 65)-(986, 155) batch id : 0
[51, 1] element, prob = 0.021973    (1046, 99)-(1107, 178) batch id : 0
[52, 1] element, prob = 0.021973    (1035, 116)-(1110, 241) batch id : 0
[53, 1] element, prob = 0.021973    (990, 166)-(1054, 269) batch id : 0
[54, 1] element, prob = 0.021973    (1065, -16)-(1192, 124) batch id : 0
[55, 1] element, prob = 0.021973    (971, 13)-(1140, 235) batch id : 0
[56, 1] element, prob = 0.021973    (980, 125)-(1108, 286) batch id : 0
[57, 1] element, prob = 0.020996    (991, 0)-(1043, 41) batch id : 0
[58, 1] element, prob = 0.020996    (1126, 0)-(1165, 42) batch id : 0
[59, 1] element, prob = 0.020996    (821, 35)-(856, 79) batch id : 0
[60, 1] element, prob = 0.020996    (910, 28)-(945, 78) batch id : 0
[61, 1] element, prob = 0.020996    (997, 41)-(1038, 86) batch id : 0
[62, 1] element, prob = 0.020996    (1068, 40)-(1103, 86) batch id : 0
[63, 1] element, prob = 0.020996    (754, 77)-(789, 125) batch id : 0
[64, 1] element, prob = 0.020996    (819, 77)-(855, 123) batch id : 0
[65, 1] element, prob = 0.020996    (913, 79)-(949, 128) batch id : 0
[66, 1] element, prob = 0.020996    (821, 112)-(861, 164) batch id : 0
[67, 1] element, prob = 0.020996    (938, 108)-(980, 173) batch id : 0
[68, 1] element, prob = 0.020996    (1149, 112)-(1189, 162) batch id : 0
[69, 1] element, prob = 0.020996    (870, 149)-(918, 215) batch id : 0
[70, 1] element, prob = 0.020996    (1033, 152)-(1070, 209) batch id : 0
[71, 1] element, prob = 0.020996    (1064, 188)-(1099, 246) batch id : 0
[72, 1] element, prob = 0.020996    (1023, 63)-(1081, 131) batch id : 0
[73, 1] element, prob = 0.020996    (1054, 58)-(1110, 126) batch id : 0
[74, 1] element, prob = 0.020996    (1041, 40)-(1125, 143) batch id : 0
[75, 1] element, prob = 0.020996    (1111, 99)-(1190, 228) batch id : 0
[76, 1] element, prob = 0.020996    (1078, 169)-(1130, 261) batch id : 0
[77, 1] element, prob = 0.020996    (950, 189)-(1072, 367) batch id : 0
[78, 1] element, prob = 0.020020    (940, 6)-(987, 56) batch id : 0
[79, 1] element, prob = 0.020020    (943, 81)-(978, 130) batch id : 0
[80, 1] element, prob = 0.020020    (878, 114)-(919, 175) batch id : 0
[81, 1] element, prob = 0.020020    (910, 110)-(950, 170) batch id : 0
[82, 1] element, prob = 0.020020    (968, 104)-(1008, 159) batch id : 0
[83, 1] element, prob = 0.020020    (1128, 141)-(1165, 204) batch id : 0
[84, 1] element, prob = 0.020020    (1147, 135)-(1188, 199) batch id : 0
[85, 1] element, prob = 0.020020    (1238, 903)-(1279, 963) batch id : 0
[86, 1] element, prob = 0.020020    (1007, -2)-(1083, 55) batch id : 0
[87, 1] element, prob = 0.020020    (1052, 161)-(1105, 266) batch id : 0
[88, 1] element, prob = 0.020020    (1061, 145)-(1140, 284) batch id : 0
[89, 1] element, prob = 0.020020    (-16, 860)-(79, 985) batch id : 0
[90, 1] element, prob = 0.020020    (866, -10)-(1029, 107) batch id : 0
[91, 1] element, prob = 0.020020    (1173, -17)-(1280, 136) batch id : 0
[92, 1] element, prob = 0.020020    (1072, 32)-(1197, 189) batch id : 0
[93, 1] element, prob = 0.020020    (1005, 179)-(1111, 359) batch id : 0
[94, 1] element, prob = 0.019043    (758, 2)-(795, 45) batch id : 0
[95, 1] element, prob = 0.019043    (968, 35)-(1011, 82) batch id : 0
[96, 1] element, prob = 0.019043    (1003, 109)-(1044, 169) batch id : 0
[97, 1] element, prob = 0.019043    (930, 148)-(973, 211) batch id : 0
[98, 1] element, prob = 0.019043    (1092, 191)-(1128, 246) batch id : 0
[99, 1] element, prob = 0.019043    (1123, 225)-(1161, 291) batch id : 0
[100, 1] element, prob = 0.019043    (1146, 19)-(1208, 109) batch id : 0
[101, 1] element, prob = 0.019043    (897, 59)-(958, 143) batch id : 0
[102, 1] element, prob = 0.019043    (962, 65)-(1023, 137) batch id : 0
[103, 1] element, prob = 0.019043    (886, 83)-(971, 209) batch id : 0
[104, 1] element, prob = 0.019043    (995, 99)-(1055, 192) batch id : 0
[105, 1] element, prob = 0.019043    (902, 135)-(963, 221) batch id : 0
[106, 1] element, prob = 0.019043    (1101, 145)-(1183, 284) batch id : 0
[107, 1] element, prob = 0.019043    (821, -17)-(946, 129) batch id : 0
[108, 1] element, prob = 0.019043    (940, -16)-(1076, 120) batch id : 0
[109, 1] element, prob = 0.019043    (819, 34)-(945, 205) batch id : 0
[110, 1] element, prob = 0.019043    (1019, 342)-(1283, 785) batch id : 0
[111, 1] element, prob = 0.018066    (1147, 7)-(1193, 54) batch id : 0
[112, 1] element, prob = 0.018066    (756, 38)-(792, 89) batch id : 0
[ INFO ] Image out_0.bmp created!
[ INFO ] Execution successful
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
➜  samples ./c/build/armv7l/Release/object_detection_sample_ssd_c -m face-detection-adas-0001.xml -d MYRIAD -i me.jpg     
[ INFO ] InferenceEngine: 
2.1.2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     me.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
         MYRIAD
         myriadPlugin version ......... 2.1
         Build ......... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image is resized from (924, 1280) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ INFO ] Start inference
[ INFO ] Processing output blobs
[0, 1] element, prob = 0.871094    (420, 202)-(745, 608) batch id : 0 WILL BE PRINTED!
[1, 1] element, prob = 0.026855    (-54, 555)-(234, 1100) batch id : 0
[2, 1] element, prob = 0.020996    (21, 925)-(239, 1340) batch id : 0
[3, 1] element, prob = 0.019043    (-18, -144)-(178, 583) batch id : 0
[4, 1] element, prob = 0.018066    (636, 315)-(757, 470) batch id : 0
[5, 1] element, prob = 0.017090    (75, 977)-(363, 1321) batch id : 0
[6, 1] element, prob = 0.016113    (291, 958)-(349, 1123) batch id : 0
[7, 1] element, prob = 0.015625    (329, 962)-(391, 1126) batch id : 0
[8, 1] element, prob = 0.015625    (832, 1121)-(937, 1293) batch id : 0
[9, 1] element, prob = 0.015625    (-27, 670)-(136, 1220) batch id : 0
[10, 1] element, prob = 0.015625    (76, 785)-(357, 1160) batch id : 0
[11, 1] element, prob = 0.015625    (204, 928)-(423, 1346) batch id : 0
[12, 1] element, prob = 0.014648    (694, 1035)-(732, 1141) batch id : 0
[13, 1] element, prob = 0.014648    (219, 40)-(388, 626) batch id : 0
[14, 1] element, prob = 0.013672    (658, 306)-(752, 405) batch id : 0
[15, 1] element, prob = 0.013672    (204, 1198)-(267, 1275) batch id : 0
[16, 1] element, prob = 0.013672    (322, 612)-(430, 793) batch id : 0
[17, 1] element, prob = 0.013672    (240, 779)-(331, 1046) batch id : 0
[18, 1] element, prob = 0.013672    (171, 245)-(424, 798) batch id : 0
[19, 1] element, prob = 0.013672    (443, 205)-(802, 1021) batch id : 0
[20, 1] element, prob = 0.012695    (694, 511)-(731, 590) batch id : 0
[21, 1] element, prob = 0.012695    (0, 893)-(24, 984) batch id : 0
[22, 1] element, prob = 0.012695    (272, 962)-(322, 1117) batch id : 0
[23, 1] element, prob = 0.012695    (268, 487)-(376, 688) batch id : 0
[24, 1] element, prob = 0.012695    (323, 490)-(422, 690) batch id : 0
[25, 1] element, prob = 0.012695    (277, 608)-(384, 795) batch id : 0
[26, 1] element, prob = 0.012695    (233, 710)-(342, 923) batch id : 0
[27, 1] element, prob = 0.012695    (256, 672)-(406, 957) batch id : 0
[28, 1] element, prob = 0.012695    (252, 799)-(540, 1176) batch id : 0
[29, 1] element, prob = 0.012695    (133, 876)-(301, 1420) batch id : 0
[30, 1] element, prob = 0.011719    (718, 466)-(757, 540) batch id : 0
[31, 1] element, prob = 0.011719    (166, 527)-(206, 594) batch id : 0
[32, 1] element, prob = 0.011719    (673, 514)-(709, 587) batch id : 0
[33, 1] element, prob = 0.011719    (-1, 943)-(22, 1038) batch id : 0
[34, 1] element, prob = 0.011719    (670, 256)-(753, 346) batch id : 0
[35, 1] element, prob = 0.011719    (705, 441)-(771, 555) batch id : 0
[36, 1] element, prob = 0.011719    (126, 625)-(200, 718) batch id : 0
[37, 1] element, prob = 0.011719    (164, 673)-(251, 747) batch id : 0
[38, 1] element, prob = 0.011719    (399, 850)-(453, 1008) batch id : 0
[39, 1] element, prob = 0.011719    (424, 851)-(476, 1009) batch id : 0
[40, 1] element, prob = 0.011719    (357, 925)-(413, 1073) batch id : 0
[41, 1] element, prob = 0.011719    (399, 919)-(454, 1074) batch id : 0
[42, 1] element, prob = 0.011719    (468, 906)-(521, 1070) batch id : 0
[43, 1] element, prob = 0.011719    (284, 1016)-(357, 1186) batch id : 0
[44, 1] element, prob = 0.011719    (269, 379)-(374, 595) batch id : 0
[45, 1] element, prob = 0.011719    (349, 563)-(485, 831) batch id : 0
[46, 1] element, prob = 0.011719    (252, 1042)-(431, 1215) batch id : 0
[47, 1] element, prob = 0.011719    (858, 771)-(987, 1518) batch id : 0
[48, 1] element, prob = 0.010742    (165, 463)-(204, 521) batch id : 0
[49, 1] element, prob = 0.010742    (185, 477)-(224, 537) batch id : 0
[50, 1] element, prob = 0.010742    (214, 477)-(250, 540) batch id : 0
[51, 1] element, prob = 0.010742    (458, 478)-(488, 536) batch id : 0
[52, 1] element, prob = 0.010742    (695, 466)-(733, 539) batch id : 0
[53, 1] element, prob = 0.010742    (460, 522)-(488, 590) batch id : 0
[54, 1] element, prob = 0.010742    (481, 523)-(509, 588) batch id : 0
[55, 1] element, prob = 0.010742    (716, 512)-(755, 592) batch id : 0
[56, 1] element, prob = 0.010742    (150, 579)-(179, 635) batch id : 0
[57, 1] element, prob = 0.010742    (171, 584)-(201, 641) batch id : 0
[58, 1] element, prob = 0.010742    (481, 582)-(510, 647) batch id : 0
[59, 1] element, prob = 0.010742    (652, 565)-(689, 650) batch id : 0
[60, 1] element, prob = 0.010742    (674, 571)-(710, 650) batch id : 0
[61, 1] element, prob = 0.010742    (694, 571)-(734, 648) batch id : 0
[62, 1] element, prob = 0.010742    (142, 640)-(188, 705) batch id : 0
[63, 1] element, prob = 0.010742    (335, 627)-(390, 695) batch id : 0
[64, 1] element, prob = 0.010742    (191, 744)-(225, 803) batch id : 0
[65, 1] element, prob = 0.010742    (370, 1005)-(399, 1078) batch id : 0
[66, 1] element, prob = 0.010742    (414, 1006)-(443, 1079) batch id : 0
[67, 1] element, prob = 0.010742    (896, 1207)-(924, 1280) batch id : 0
[68, 1] element, prob = 0.010742    (86, 386)-(160, 513) batch id : 0
[69, 1] element, prob = 0.010742    (106, 477)-(214, 630) batch id : 0
[70, 1] element, prob = 0.010742    (151, 501)-(221, 610) batch id : 0
[71, 1] element, prob = 0.010742    (428, 478)-(479, 615) batch id : 0
[72, 1] element, prob = 0.010742    (450, 478)-(506, 611) batch id : 0
[73, 1] element, prob = 0.010742    (647, 498)-(728, 605) batch id : 0
[74, 1] element, prob = 0.010742    (148, 556)-(220, 658) batch id : 0
[75, 1] element, prob = 0.010742    (425, 557)-(481, 671) batch id : 0
[76, 1] element, prob = 0.010742    (437, 510)-(516, 700) batch id : 0
[77, 1] element, prob = 0.010742    (463, 554)-(527, 674) batch id : 0
[78, 1] element, prob = 0.010742    (633, 551)-(704, 664) batch id : 0
[79, 1] element, prob = 0.010742    (280, 628)-(367, 711) batch id : 0
[80, 1] element, prob = 0.010742    (416, 612)-(488, 735) batch id : 0
[81, 1] element, prob = 0.010742    (155, 706)-(260, 850) batch id : 0
[82, 1] element, prob = 0.010742    (55, 828)-(137, 932) batch id : 0
[83, 1] element, prob = 0.010742    (451, 813)-(498, 960) batch id : 0
[84, 1] element, prob = 0.010742    (310, 914)-(371, 1076) batch id : 0
[85, 1] element, prob = 0.010742    (447, 896)-(500, 1065) batch id : 0
[86, 1] element, prob = 0.010742    (641, 966)-(706, 1121) batch id : 0
[87, 1] element, prob = 0.010742    (203, -95)-(283, 182) batch id : 0
[88, 1] element, prob = 0.010742    (164, 86)-(233, 410) batch id : 0
[89, 1] element, prob = 0.010742    (188, 498)-(307, 688) batch id : 0
[90, 1] element, prob = 0.010742    (205, 445)-(357, 715) batch id : 0
[91, 1] element, prob = 0.010742    (-22, 576)-(71, 832) batch id : 0
[92, 1] element, prob = 0.010742    (230, 603)-(342, 791) batch id : 0
[93, 1] element, prob = 0.010742    (318, 698)-(433, 909) batch id : 0
[94, 1] element, prob = 0.010742    (273, 784)-(390, 1021) batch id : 0
[95, 1] element, prob = 0.010742    (175, 740)-(444, 1237) batch id : 0
[ INFO ] Image out_0.bmp created!
[ INFO ] Execution successful

3.3.2 C++

For NCS 2

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
➜  samples ./cpp/build/armv7l/Release/object_detection_sample_ssd -m face-detection-adas-0001.xml -d MYRIAD -i me.jpg 
[ INFO ] InferenceEngine: 
	API version ............ 2.1
	Build .................. 2020.3.0-3467-15f2c61a-releases/2020/3
	Description ....... API
Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     me.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
	MYRIAD
	myriadPlugin version ......... 2.1
	Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ WARNING ] Image is resized from (924, 1280) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Start inference
[ INFO ] Processing output blobs
[0,1] element, prob = 0.870117    (421,202)-(745,608) batch id : 0 WILL BE PRINTED!
[1,1] element, prob = 0.0268555    (-55,556)-(234,1100) batch id : 0
[2,1] element, prob = 0.0219727    (20,925)-(239,1340) batch id : 0
[3,1] element, prob = 0.0200195    (-18,-144)-(179,581) batch id : 0
[4,1] element, prob = 0.0180664    (636,315)-(757,471) batch id : 0
[5,1] element, prob = 0.0170898    (-27,672)-(137,1217) batch id : 0
[6,1] element, prob = 0.0170898    (74,976)-(364,1322) batch id : 0
[7,1] element, prob = 0.0161133    (291,958)-(349,1122) batch id : 0
[8,1] element, prob = 0.0161133    (329,963)-(391,1126) batch id : 0
[9,1] element, prob = 0.0161133    (203,928)-(423,1346) batch id : 0
[10,1] element, prob = 0.015625    (694,1034)-(732,1140) batch id : 0
[11,1] element, prob = 0.015625    (832,1121)-(937,1295) batch id : 0
[12,1] element, prob = 0.015625    (219,39)-(388,627) batch id : 0
[13,1] element, prob = 0.015625    (76,785)-(357,1160) batch id : 0
[14,1] element, prob = 0.0146484    (355,616)-(475,778) batch id : 0
[15,1] element, prob = 0.0146484    (240,779)-(331,1048) batch id : 0
[16,1] element, prob = 0.0136719    (0,893)-(24,984) batch id : 0
[17,1] element, prob = 0.0136719    (658,306)-(752,405) batch id : 0
[18,1] element, prob = 0.0136719    (204,1198)-(267,1275) batch id : 0
[19,1] element, prob = 0.0136719    (322,611)-(430,794) batch id : 0
[20,1] element, prob = 0.0136719    (171,244)-(423,798) batch id : 0
[21,1] element, prob = 0.0136719    (252,798)-(539,1177) batch id : 0
[22,1] element, prob = 0.0136719    (133,875)-(301,1420) batch id : 0
[23,1] element, prob = 0.0136719    (443,202)-(802,1022) batch id : 0
[24,1] element, prob = 0.0126953    (718,466)-(757,540) batch id : 0
[25,1] element, prob = 0.0126953    (166,527)-(206,594) batch id : 0
[26,1] element, prob = 0.0126953    (694,511)-(731,590) batch id : 0
[27,1] element, prob = 0.0126953    (705,441)-(771,555) batch id : 0
[28,1] element, prob = 0.0126953    (272,962)-(322,1117) batch id : 0
[29,1] element, prob = 0.0126953    (268,486)-(376,688) batch id : 0
[30,1] element, prob = 0.0126953    (323,489)-(422,690) batch id : 0
[31,1] element, prob = 0.0126953    (277,607)-(384,796) batch id : 0
[32,1] element, prob = 0.0126953    (233,709)-(342,924) batch id : 0
[33,1] element, prob = 0.0126953    (256,672)-(406,957) batch id : 0
[34,1] element, prob = 0.0126953    (252,1043)-(431,1214) batch id : 0
[35,1] element, prob = 0.0117188    (695,466)-(733,539) batch id : 0
[36,1] element, prob = 0.0117188    (673,514)-(709,587) batch id : 0
[37,1] element, prob = 0.0117188    (370,1005)-(399,1078) batch id : 0
[38,1] element, prob = 0.0117188    (670,256)-(753,346) batch id : 0
[39,1] element, prob = 0.0117188    (151,500)-(221,610) batch id : 0
[40,1] element, prob = 0.0117188    (447,489)-(511,608) batch id : 0
[41,1] element, prob = 0.0117188    (126,625)-(200,719) batch id : 0
[42,1] element, prob = 0.0117188    (299,626)-(387,709) batch id : 0
[43,1] element, prob = 0.0117188    (164,673)-(251,747) batch id : 0
[44,1] element, prob = 0.0117188    (398,850)-(453,1008) batch id : 0
[45,1] element, prob = 0.0117188    (425,851)-(476,1008) batch id : 0
[46,1] element, prob = 0.0117188    (310,915)-(371,1077) batch id : 0
[47,1] element, prob = 0.0117188    (357,925)-(413,1073) batch id : 0
[48,1] element, prob = 0.0117188    (378,923)-(435,1074) batch id : 0
[49,1] element, prob = 0.0117188    (468,906)-(521,1070) batch id : 0
[50,1] element, prob = 0.0117188    (284,1016)-(357,1188) batch id : 0
[51,1] element, prob = 0.0117188    (203,-95)-(283,182) batch id : 0
[52,1] element, prob = 0.0117188    (269,378)-(374,595) batch id : 0
[53,1] element, prob = 0.0117188    (-22,575)-(71,831) batch id : 0
[54,1] element, prob = 0.0117188    (318,698)-(433,911) batch id : 0
[55,1] element, prob = 0.0117188    (221,716)-(395,1258) batch id : 0
[56,1] element, prob = 0.0117188    (859,771)-(987,1518) batch id : 0
[57,1] element, prob = 0.0107422    (145,466)-(185,527) batch id : 0
[58,1] element, prob = 0.0107422    (166,463)-(204,521) batch id : 0
[59,1] element, prob = 0.0107422    (185,477)-(224,537) batch id : 0
[60,1] element, prob = 0.0107422    (214,477)-(250,540) batch id : 0
[61,1] element, prob = 0.0107422    (458,478)-(488,536) batch id : 0
[62,1] element, prob = 0.0107422    (151,537)-(177,594) batch id : 0
[63,1] element, prob = 0.0107422    (460,522)-(488,590) batch id : 0
[64,1] element, prob = 0.0107422    (481,523)-(509,588) batch id : 0
[65,1] element, prob = 0.0107422    (716,512)-(755,592) batch id : 0
[66,1] element, prob = 0.0107422    (129,581)-(158,641) batch id : 0
[67,1] element, prob = 0.0107422    (150,579)-(179,635) batch id : 0
[68,1] element, prob = 0.0107422    (171,583)-(201,641) batch id : 0
[69,1] element, prob = 0.0107422    (457,564)-(491,658) batch id : 0
[70,1] element, prob = 0.0107422    (481,581)-(510,647) batch id : 0
[71,1] element, prob = 0.0107422    (652,565)-(689,650) batch id : 0
[72,1] element, prob = 0.0107422    (674,571)-(710,650) batch id : 0
[73,1] element, prob = 0.0107422    (694,571)-(734,649) batch id : 0
[74,1] element, prob = 0.0107422    (143,640)-(188,705) batch id : 0
[75,1] element, prob = 0.0107422    (191,743)-(224,805) batch id : 0
[76,1] element, prob = 0.0107422    (437,843)-(465,920) batch id : 0
[77,1] element, prob = 0.0107422    (-1,943)-(22,1037) batch id : 0
[78,1] element, prob = 0.0107422    (414,1007)-(443,1080) batch id : 0
[79,1] element, prob = 0.0107422    (896,1206)-(924,1279) batch id : 0
[80,1] element, prob = 0.0107422    (678,200)-(743,294) batch id : 0
[81,1] element, prob = 0.0107422    (70,374)-(190,512) batch id : 0
[82,1] element, prob = 0.0107422    (422,429)-(485,563) batch id : 0
[83,1] element, prob = 0.0107422    (106,477)-(213,629) batch id : 0
[84,1] element, prob = 0.0107422    (428,477)-(479,615) batch id : 0
[85,1] element, prob = 0.0107422    (648,498)-(727,605) batch id : 0
[86,1] element, prob = 0.0107422    (703,490)-(769,614) batch id : 0
[87,1] element, prob = 0.0107422    (16,531)-(150,712) batch id : 0
[88,1] element, prob = 0.0107422    (87,546)-(150,685) batch id : 0
[89,1] element, prob = 0.0107422    (127,552)-(197,653) batch id : 0
[90,1] element, prob = 0.0107422    (425,557)-(481,671) batch id : 0
[91,1] element, prob = 0.0107422    (437,509)-(516,700) batch id : 0
[92,1] element, prob = 0.0107422    (467,549)-(524,678) batch id : 0
[93,1] element, prob = 0.0107422    (633,551)-(704,664) batch id : 0
[94,1] element, prob = 0.0107422    (416,611)-(487,735) batch id : 0
[95,1] element, prob = 0.0107422    (466,602)-(530,734) batch id : 0
[96,1] element, prob = 0.0107422    (124,673)-(198,759) batch id : 0
[97,1] element, prob = 0.0107422    (156,705)-(260,849) batch id : 0
[98,1] element, prob = 0.0107422    (54,828)-(137,933) batch id : 0
[99,1] element, prob = 0.0107422    (451,812)-(498,960) batch id : 0
[100,1] element, prob = 0.0107422    (470,843)-(520,1004) batch id : 0
[101,1] element, prob = 0.0107422    (447,897)-(500,1066) batch id : 0
[102,1] element, prob = 0.0107422    (641,966)-(706,1122) batch id : 0
[103,1] element, prob = 0.0107422    (164,85)-(233,410) batch id : 0
[104,1] element, prob = 0.0107422    (188,497)-(307,688) batch id : 0
[105,1] element, prob = 0.0107422    (205,445)-(357,715) batch id : 0
[106,1] element, prob = 0.0107422    (230,602)-(341,791) batch id : 0
[107,1] element, prob = 0.0107422    (274,785)-(390,1021) batch id : 0
[108,1] element, prob = 0.0107422    (26,709)-(209,1239) batch id : 0
[109,1] element, prob = 0.0107422    (309,868)-(485,1410) batch id : 0
[110,1] element, prob = 0.0107422    (229,981)-(575,1306) batch id : 0
[111,1] element, prob = 0.0107422    (29,141)-(518,883) batch id : 0
[ INFO ] Image out_0.bmp created!
[ INFO ] Execution successful

[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
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
➜  samples ./cpp/build/armv7l/Release/object_detection_sample_ssd -m face-detection-adas-0001.xml -d MYRIAD -i parents.jpg
[ INFO ] InferenceEngine: 
	API version ............ 2.1
	Build .................. 2020.3.0-3467-15f2c61a-releases/2020/3
	Description ....... API
Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     parents.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
	MYRIAD
	myriadPlugin version ......... 2.1
	Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ WARNING ] Image is resized from (1280, 960) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Start inference
[ INFO ] Processing output blobs
[0,1] element, prob = 1    (826,366)-(1026,644) batch id : 0 WILL BE PRINTED!
[1,1] element, prob = 0.996582    (539,173)-(693,429) batch id : 0 WILL BE PRINTED!
[2,1] element, prob = 0.539062    (1094,47)-(1135,96) batch id : 0 WILL BE PRINTED!
[3,1] element, prob = 0.212402    (848,22)-(886,70) batch id : 0
[4,1] element, prob = 0.0527344    (7,783)-(151,957) batch id : 0
[5,1] element, prob = 0.0395508    (1033,78)-(1070,122) batch id : 0
[6,1] element, prob = 0.034668    (1123,75)-(1158,122) batch id : 0
[7,1] element, prob = 0.0322266    (1091,109)-(1130,171) batch id : 0
[8,1] element, prob = 0.0322266    (1086,14)-(1138,85) batch id : 0
[9,1] element, prob = 0.0302734    (1109,43)-(1150,93) batch id : 0
[10,1] element, prob = 0.0292969    (1151,83)-(1190,131) batch id : 0
[11,1] element, prob = 0.0273438    (1030,2)-(1073,43) batch id : 0
[12,1] element, prob = 0.0273438    (1091,6)-(1133,55) batch id : 0
[13,1] element, prob = 0.0273438    (1064,73)-(1098,121) batch id : 0
[14,1] element, prob = 0.0273438    (1058,113)-(1096,168) batch id : 0
[15,1] element, prob = 0.0273438    (1143,74)-(1211,165) batch id : 0
[16,1] element, prob = 0.0268555    (876,73)-(915,118) batch id : 0
[17,1] element, prob = 0.0268555    (1047,-3)-(1110,59) batch id : 0
[18,1] element, prob = 0.0268555    (849,13)-(1030,236) batch id : 0
[19,1] element, prob = 0.0258789    (941,22)-(981,74) batch id : 0
[20,1] element, prob = 0.0258789    (1030,35)-(1074,89) batch id : 0
[21,1] element, prob = 0.0258789    (845,112)-(886,167) batch id : 0
[22,1] element, prob = 0.0258789    (1036,111)-(1076,171) batch id : 0
[23,1] element, prob = 0.0258789    (1113,-1)-(1175,59) batch id : 0
[24,1] element, prob = 0.0258789    (1002,40)-(1093,151) batch id : 0
[25,1] element, prob = 0.0249023    (1059,0)-(1098,43) batch id : 0
[26,1] element, prob = 0.0249023    (876,29)-(918,79) batch id : 0
[27,1] element, prob = 0.0249023    (1002,69)-(1048,123) batch id : 0
[28,1] element, prob = 0.0249023    (1133,104)-(1170,163) batch id : 0
[29,1] element, prob = 0.0249023    (1056,145)-(1094,208) batch id : 0
[30,1] element, prob = 0.0249023    (861,2)-(918,63) batch id : 0
[31,1] element, prob = 0.0239258    (966,1)-(1018,45) batch id : 0
[32,1] element, prob = 0.0239258    (1098,63)-(1131,109) batch id : 0
[33,1] element, prob = 0.0239258    (1010,84)-(1092,204) batch id : 0
[34,1] element, prob = 0.0239258    (1022,139)-(1076,224) batch id : 0
[35,1] element, prob = 0.0239258    (943,34)-(1073,208) batch id : 0
[36,1] element, prob = 0.0229492    (811,3)-(851,48) batch id : 0
[37,1] element, prob = 0.0229492    (1154,27)-(1198,79) batch id : 0
[38,1] element, prob = 0.0229492    (1000,155)-(1046,226) batch id : 0
[39,1] element, prob = 0.0229492    (1086,145)-(1125,208) batch id : 0
[40,1] element, prob = 0.0229492    (958,20)-(1030,84) batch id : 0
[41,1] element, prob = 0.0229492    (934,65)-(986,155) batch id : 0
[42,1] element, prob = 0.0229492    (1045,99)-(1106,178) batch id : 0
[43,1] element, prob = 0.0229492    (983,-9)-(1141,105) batch id : 0
[44,1] element, prob = 0.0229492    (1065,-16)-(1192,124) batch id : 0
[45,1] element, prob = 0.0229492    (922,81)-(1087,304) batch id : 0
[46,1] element, prob = 0.0219727    (796,3)-(834,47) batch id : 0
[47,1] element, prob = 0.0219727    (911,4)-(953,55) batch id : 0
[48,1] element, prob = 0.0219727    (797,29)-(835,79) batch id : 0
[49,1] element, prob = 0.0219727    (821,34)-(855,81) batch id : 0
[50,1] element, prob = 0.0219727    (910,28)-(945,78) batch id : 0
[51,1] element, prob = 0.0219727    (755,77)-(790,125) batch id : 0
[52,1] element, prob = 0.0219727    (969,80)-(1008,128) batch id : 0
[53,1] element, prob = 0.0219727    (938,108)-(980,173) batch id : 0
[54,1] element, prob = 0.0219727    (1123,186)-(1161,252) batch id : 0
[55,1] element, prob = 0.0219727    (985,28)-(1056,105) batch id : 0
[56,1] element, prob = 0.0219727    (845,56)-(898,135) batch id : 0
[57,1] element, prob = 0.0219727    (1023,62)-(1080,131) batch id : 0
[58,1] element, prob = 0.0219727    (1041,40)-(1125,143) batch id : 0
[59,1] element, prob = 0.0219727    (1035,116)-(1110,242) batch id : 0
[60,1] element, prob = 0.0219727    (990,166)-(1054,269) batch id : 0
[61,1] element, prob = 0.0219727    (1078,169)-(1130,261) batch id : 0
[62,1] element, prob = 0.0219727    (971,13)-(1140,235) batch id : 0
[63,1] element, prob = 0.0219727    (980,125)-(1108,286) batch id : 0
[64,1] element, prob = 0.0209961    (940,5)-(987,55) batch id : 0
[65,1] element, prob = 0.0209961    (991,0)-(1043,41) batch id : 0
[66,1] element, prob = 0.0209961    (1126,0)-(1165,42) batch id : 0
[67,1] element, prob = 0.0209961    (998,41)-(1039,86) batch id : 0
[68,1] element, prob = 0.0209961    (1068,40)-(1103,86) batch id : 0
[69,1] element, prob = 0.0209961    (821,77)-(856,123) batch id : 0
[70,1] element, prob = 0.0209961    (913,80)-(949,128) batch id : 0
[71,1] element, prob = 0.0209961    (821,112)-(861,164) batch id : 0
[72,1] element, prob = 0.0209961    (878,113)-(919,175) batch id : 0
[73,1] element, prob = 0.0209961    (910,110)-(950,170) batch id : 0
[74,1] element, prob = 0.0209961    (1149,111)-(1189,163) batch id : 0
[75,1] element, prob = 0.0209961    (870,149)-(918,215) batch id : 0
[76,1] element, prob = 0.0209961    (1033,152)-(1070,209) batch id : 0
[77,1] element, prob = 0.0209961    (1128,141)-(1165,204) batch id : 0
[78,1] element, prob = 0.0209961    (1147,135)-(1188,199) batch id : 0
[79,1] element, prob = 0.0209961    (1064,188)-(1099,246) batch id : 0
[80,1] element, prob = 0.0209961    (1238,903)-(1279,963) batch id : 0
[81,1] element, prob = 0.0209961    (1146,19)-(1208,109) batch id : 0
[82,1] element, prob = 0.0209961    (1054,58)-(1110,126) batch id : 0
[83,1] element, prob = 0.0209961    (1111,99)-(1190,228) batch id : 0
[84,1] element, prob = 0.0209961    (1052,161)-(1105,265) batch id : 0
[85,1] element, prob = 0.0209961    (1060,145)-(1139,284) batch id : 0
[86,1] element, prob = 0.0209961    (-16,860)-(79,985) batch id : 0
[87,1] element, prob = 0.0209961    (1174,-17)-(1281,136) batch id : 0
[88,1] element, prob = 0.0209961    (1073,32)-(1198,190) batch id : 0
[89,1] element, prob = 0.0209961    (950,188)-(1072,366) batch id : 0
[90,1] element, prob = 0.0200195    (943,81)-(978,130) batch id : 0
[91,1] element, prob = 0.0200195    (968,104)-(1008,159) batch id : 0
[92,1] element, prob = 0.0200195    (930,148)-(973,211) batch id : 0
[93,1] element, prob = 0.0200195    (1092,191)-(1128,246) batch id : 0
[94,1] element, prob = 0.0200195    (1124,225)-(1161,291) batch id : 0
[95,1] element, prob = 0.0200195    (1007,-2)-(1083,55) batch id : 0
[96,1] element, prob = 0.0200195    (890,60)-(960,138) batch id : 0
[97,1] element, prob = 0.0200195    (962,65)-(1023,137) batch id : 0
[98,1] element, prob = 0.0200195    (1101,145)-(1183,284) batch id : 0
[99,1] element, prob = 0.0200195    (866,-10)-(1029,107) batch id : 0
[100,1] element, prob = 0.0200195    (1005,179)-(1111,360) batch id : 0
[101,1] element, prob = 0.0200195    (1019,342)-(1283,785) batch id : 0
[102,1] element, prob = 0.019043    (758,2)-(795,45) batch id : 0
[103,1] element, prob = 0.019043    (756,38)-(792,89) batch id : 0
[104,1] element, prob = 0.019043    (968,36)-(1011,82) batch id : 0
[105,1] element, prob = 0.019043    (729,77)-(763,127) batch id : 0
[106,1] element, prob = 0.019043    (1003,109)-(1044,169) batch id : 0
[107,1] element, prob = 0.019043    (1030,190)-(1066,242) batch id : 0
[108,1] element, prob = 0.019043    (887,83)-(971,209) batch id : 0
[109,1] element, prob = 0.019043    (995,99)-(1055,192) batch id : 0
[110,1] element, prob = 0.019043    (902,135)-(963,221) batch id : 0
[ INFO ] Image out_0.bmp created!
object_detection_sample_ssd: ../../libusb/io.c:2116: handle_events: Assertion `ctx->pollfds_cnt >= internal_nfds' failed.
Aborted

For NCS 1

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
➜  samples ./cpp/build/armv7l/Release/object_detection_sample_ssd -m face-detection-adas-0001.xml -d MYRIAD -i me.jpg     
[ INFO ] InferenceEngine: 
	API version ............ 2.1
	Build .................. 2020.3.0-3467-15f2c61a-releases/2020/3
	Description ....... API
Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     me.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
	MYRIAD
	myriadPlugin version ......... 2.1
	Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ WARNING ] Image is resized from (924, 1280) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Start inference
[ INFO ] Processing output blobs
object_detection_sample_ssd: ../../libusb/io.c:2116: handle_events: Assertion `ctx->pollfds_cnt >= internal_nfds' failed.
Aborted
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
➜  samples ./cpp/build/armv7l/Release/object_detection_sample_ssd -m face-detection-adas-0001.xml -d MYRIAD -i parents.jpg
[ INFO ] InferenceEngine: 
	API version ............ 2.1
	Build .................. 2020.3.0-3467-15f2c61a-releases/2020/3
	Description ....... API
Parsing input parameters
[ INFO ] Files were added: 1
[ INFO ]     parents.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Device info: 
	MYRIAD
	myriadPlugin version ......... 2.1
	Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Create infer request
[ WARNING ] Image is resized from (1280, 960) to (672, 384)
[ INFO ] Batch size is 1
[ INFO ] Start inference
[ INFO ] Processing output blobs
object_detection_sample_ssd: ../../libusb/io.c:2116: handle_events: Assertion `ctx->pollfds_cnt >= internal_nfds' failed.
Aborted

3.3.3 Python

For NCS 2

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
➜  samples python ./python/object_detection_sample_ssd/object_detection_sample_ssd.py -m face-detection-adas-0001.xml -d MYRIAD -i me.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Device info:
        MYRIAD
        MKLDNNPlugin version ......... 2.1
        Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
inputs number: 1
input shape: [1, 3, 384, 672]
input key: data
[ INFO ] File was added: 
[ INFO ]         me.jpg
[ WARNING ] Image me.jpg is resized from (384, 672) to (384, 672)
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Creating infer request and starting inference
[ INFO ] Processing output blobs
[0,1] element, prob = 0.870117    (421,202)-(745,608) batch id : 0 WILL BE PRINTED!
[1,1] element, prob = 0.0268555    (-55,556)-(234,1100) batch id : 0
[2,1] element, prob = 0.0219727    (20,925)-(239,1340) batch id : 0
[3,1] element, prob = 0.0200195    (-18,-144)-(179,581) batch id : 0
[4,1] element, prob = 0.0180664    (636,315)-(757,471) batch id : 0
[5,1] element, prob = 0.0170898    (-27,672)-(137,1217) batch id : 0
[6,1] element, prob = 0.0170898    (74,976)-(364,1322) batch id : 0
[7,1] element, prob = 0.0161133    (291,958)-(349,1122) batch id : 0
[8,1] element, prob = 0.0161133    (329,963)-(391,1126) batch id : 0
[9,1] element, prob = 0.0161133    (203,928)-(423,1346) batch id : 0
[10,1] element, prob = 0.015625    (694,1034)-(732,1140) batch id : 0
[11,1] element, prob = 0.015625    (832,1121)-(937,1295) batch id : 0
[12,1] element, prob = 0.015625    (219,39)-(388,627) batch id : 0
[13,1] element, prob = 0.015625    (76,785)-(357,1160) batch id : 0
[14,1] element, prob = 0.0146484    (355,616)-(475,778) batch id : 0
[15,1] element, prob = 0.0146484    (240,779)-(331,1048) batch id : 0
[16,1] element, prob = 0.0136719    (0,893)-(24,984) batch id : 0
[17,1] element, prob = 0.0136719    (658,306)-(752,405) batch id : 0
[18,1] element, prob = 0.0136719    (204,1198)-(267,1275) batch id : 0
[19,1] element, prob = 0.0136719    (322,611)-(430,794) batch id : 0
[20,1] element, prob = 0.0136719    (171,244)-(423,798) batch id : 0
[21,1] element, prob = 0.0136719    (252,798)-(539,1177) batch id : 0
[22,1] element, prob = 0.0136719    (133,875)-(301,1420) batch id : 0
[23,1] element, prob = 0.0136719    (443,202)-(802,1022) batch id : 0
[24,1] element, prob = 0.0126953    (718,466)-(757,540) batch id : 0
[25,1] element, prob = 0.0126953    (166,527)-(206,594) batch id : 0
[26,1] element, prob = 0.0126953    (694,511)-(731,590) batch id : 0
[27,1] element, prob = 0.0126953    (705,441)-(771,555) batch id : 0
[28,1] element, prob = 0.0126953    (272,962)-(322,1117) batch id : 0
[29,1] element, prob = 0.0126953    (268,486)-(376,688) batch id : 0
[30,1] element, prob = 0.0126953    (323,489)-(422,690) batch id : 0
[31,1] element, prob = 0.0126953    (277,607)-(384,796) batch id : 0
[32,1] element, prob = 0.0126953    (233,709)-(342,924) batch id : 0
[33,1] element, prob = 0.0126953    (256,672)-(406,957) batch id : 0
[34,1] element, prob = 0.0126953    (252,1043)-(431,1214) batch id : 0
[35,1] element, prob = 0.0117188    (695,466)-(733,539) batch id : 0
[36,1] element, prob = 0.0117188    (673,514)-(709,587) batch id : 0
[37,1] element, prob = 0.0117188    (370,1005)-(399,1078) batch id : 0
[38,1] element, prob = 0.0117188    (670,256)-(753,346) batch id : 0
[39,1] element, prob = 0.0117188    (151,500)-(221,610) batch id : 0
[40,1] element, prob = 0.0117188    (447,489)-(511,608) batch id : 0
[41,1] element, prob = 0.0117188    (126,625)-(200,719) batch id : 0
[42,1] element, prob = 0.0117188    (299,626)-(387,709) batch id : 0
[43,1] element, prob = 0.0117188    (164,673)-(251,747) batch id : 0
[44,1] element, prob = 0.0117188    (398,850)-(453,1008) batch id : 0
[45,1] element, prob = 0.0117188    (425,851)-(476,1008) batch id : 0
[46,1] element, prob = 0.0117188    (310,915)-(371,1077) batch id : 0
[47,1] element, prob = 0.0117188    (357,925)-(413,1073) batch id : 0
[48,1] element, prob = 0.0117188    (378,923)-(435,1074) batch id : 0
[49,1] element, prob = 0.0117188    (468,906)-(521,1070) batch id : 0
[50,1] element, prob = 0.0117188    (284,1016)-(357,1188) batch id : 0
[51,1] element, prob = 0.0117188    (203,-95)-(283,182) batch id : 0
[52,1] element, prob = 0.0117188    (269,378)-(374,595) batch id : 0
[53,1] element, prob = 0.0117188    (-22,575)-(71,831) batch id : 0
[54,1] element, prob = 0.0117188    (318,698)-(433,911) batch id : 0
[55,1] element, prob = 0.0117188    (221,716)-(395,1258) batch id : 0
[56,1] element, prob = 0.0117188    (859,771)-(987,1518) batch id : 0
[57,1] element, prob = 0.0107422    (145,466)-(185,527) batch id : 0
[58,1] element, prob = 0.0107422    (166,463)-(204,521) batch id : 0
[59,1] element, prob = 0.0107422    (185,477)-(224,537) batch id : 0
[60,1] element, prob = 0.0107422    (214,477)-(250,540) batch id : 0
[61,1] element, prob = 0.0107422    (458,478)-(488,536) batch id : 0
[62,1] element, prob = 0.0107422    (151,537)-(177,594) batch id : 0
[63,1] element, prob = 0.0107422    (460,522)-(488,590) batch id : 0
[64,1] element, prob = 0.0107422    (481,523)-(509,588) batch id : 0
[65,1] element, prob = 0.0107422    (716,512)-(755,592) batch id : 0
[66,1] element, prob = 0.0107422    (129,581)-(158,641) batch id : 0
[67,1] element, prob = 0.0107422    (150,579)-(179,635) batch id : 0
[68,1] element, prob = 0.0107422    (171,583)-(201,641) batch id : 0
[69,1] element, prob = 0.0107422    (457,564)-(491,658) batch id : 0
[70,1] element, prob = 0.0107422    (481,581)-(510,647) batch id : 0
[71,1] element, prob = 0.0107422    (652,565)-(689,650) batch id : 0
[72,1] element, prob = 0.0107422    (674,571)-(710,650) batch id : 0
[73,1] element, prob = 0.0107422    (694,571)-(734,649) batch id : 0
[74,1] element, prob = 0.0107422    (143,640)-(188,705) batch id : 0
[75,1] element, prob = 0.0107422    (191,743)-(224,805) batch id : 0
[76,1] element, prob = 0.0107422    (437,843)-(465,920) batch id : 0
[77,1] element, prob = 0.0107422    (-1,943)-(22,1037) batch id : 0
[78,1] element, prob = 0.0107422    (414,1007)-(443,1080) batch id : 0
[79,1] element, prob = 0.0107422    (896,1206)-(924,1279) batch id : 0
[80,1] element, prob = 0.0107422    (678,200)-(743,294) batch id : 0
[81,1] element, prob = 0.0107422    (70,374)-(190,512) batch id : 0
[82,1] element, prob = 0.0107422    (422,429)-(485,563) batch id : 0
[83,1] element, prob = 0.0107422    (106,477)-(213,629) batch id : 0
[84,1] element, prob = 0.0107422    (428,477)-(479,615) batch id : 0
[85,1] element, prob = 0.0107422    (648,498)-(727,605) batch id : 0
[86,1] element, prob = 0.0107422    (703,490)-(769,614) batch id : 0
[87,1] element, prob = 0.0107422    (16,531)-(150,712) batch id : 0
[88,1] element, prob = 0.0107422    (87,546)-(150,685) batch id : 0
[89,1] element, prob = 0.0107422    (127,552)-(197,653) batch id : 0
[90,1] element, prob = 0.0107422    (425,557)-(481,671) batch id : 0
[91,1] element, prob = 0.0107422    (437,509)-(516,700) batch id : 0
[92,1] element, prob = 0.0107422    (467,549)-(524,678) batch id : 0
[93,1] element, prob = 0.0107422    (633,551)-(704,664) batch id : 0
[94,1] element, prob = 0.0107422    (416,611)-(487,735) batch id : 0
[95,1] element, prob = 0.0107422    (466,602)-(530,734) batch id : 0
[96,1] element, prob = 0.0107422    (124,673)-(198,759) batch id : 0
[97,1] element, prob = 0.0107422    (156,705)-(260,849) batch id : 0
[98,1] element, prob = 0.0107422    (54,828)-(137,933) batch id : 0
[99,1] element, prob = 0.0107422    (451,812)-(498,960) batch id : 0
[100,1] element, prob = 0.0107422    (470,843)-(520,1004) batch id : 0
[101,1] element, prob = 0.0107422    (447,897)-(500,1066) batch id : 0
[102,1] element, prob = 0.0107422    (641,966)-(706,1122) batch id : 0
[103,1] element, prob = 0.0107422    (164,85)-(233,410) batch id : 0
[104,1] element, prob = 0.0107422    (188,497)-(307,688) batch id : 0
[105,1] element, prob = 0.0107422    (205,445)-(357,715) batch id : 0
[106,1] element, prob = 0.0107422    (230,602)-(341,791) batch id : 0
[107,1] element, prob = 0.0107422    (274,785)-(390,1021) batch id : 0
[108,1] element, prob = 0.0107422    (26,709)-(209,1239) batch id : 0
[109,1] element, prob = 0.0107422    (309,868)-(485,1410) batch id : 0
[110,1] element, prob = 0.0107422    (229,981)-(575,1306) batch id : 0
[111,1] element, prob = 0.0107422    (29,141)-(518,883) batch id : 0
[ INFO ] Image out.bmp created!
[ INFO ] Execution successful

[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
corrupted double-linked list
Aborted
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
➜  samples python ./python/object_detection_sample_ssd/object_detection_sample_ssd.py -m face-detection-adas-0001.xml -d MYRIAD -i parents.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Device info:
        MYRIAD
        MKLDNNPlugin version ......... 2.1
        Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
inputs number: 1
input shape: [1, 3, 384, 672]
input key: data
[ INFO ] File was added: 
[ INFO ]         parents.jpg
[ WARNING ] Image parents.jpg is resized from (384, 672) to (384, 672)
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Creating infer request and starting inference
[ INFO ] Processing output blobs
[0,1] element, prob = 1.0    (826,366)-(1026,644) batch id : 0 WILL BE PRINTED!
[1,1] element, prob = 0.996582    (539,173)-(693,429) batch id : 0 WILL BE PRINTED!
[2,1] element, prob = 0.539062    (1094,47)-(1135,96) batch id : 0 WILL BE PRINTED!
[3,1] element, prob = 0.212402    (848,22)-(886,70) batch id : 0
[4,1] element, prob = 0.0527344    (7,783)-(151,957) batch id : 0
[5,1] element, prob = 0.0395508    (1033,78)-(1070,122) batch id : 0
[6,1] element, prob = 0.034668    (1123,75)-(1158,122) batch id : 0
[7,1] element, prob = 0.0322266    (1091,109)-(1130,171) batch id : 0
[8,1] element, prob = 0.0322266    (1086,14)-(1138,85) batch id : 0
[9,1] element, prob = 0.0302734    (1109,43)-(1150,93) batch id : 0
[10,1] element, prob = 0.0292969    (1151,83)-(1190,131) batch id : 0
[11,1] element, prob = 0.0273438    (1030,2)-(1073,43) batch id : 0
[12,1] element, prob = 0.0273438    (1091,6)-(1133,55) batch id : 0
[13,1] element, prob = 0.0273438    (1064,73)-(1098,121) batch id : 0
[14,1] element, prob = 0.0273438    (1058,113)-(1096,168) batch id : 0
[15,1] element, prob = 0.0273438    (1143,74)-(1211,165) batch id : 0
[16,1] element, prob = 0.0268555    (876,73)-(915,118) batch id : 0
[17,1] element, prob = 0.0268555    (1047,-3)-(1110,59) batch id : 0
[18,1] element, prob = 0.0268555    (849,13)-(1030,236) batch id : 0
[19,1] element, prob = 0.0258789    (941,22)-(981,74) batch id : 0
[20,1] element, prob = 0.0258789    (1030,35)-(1074,89) batch id : 0
[21,1] element, prob = 0.0258789    (845,112)-(886,167) batch id : 0
[22,1] element, prob = 0.0258789    (1036,111)-(1076,171) batch id : 0
[23,1] element, prob = 0.0258789    (1113,-1)-(1175,59) batch id : 0
[24,1] element, prob = 0.0258789    (1002,40)-(1093,151) batch id : 0
[25,1] element, prob = 0.0249023    (1059,0)-(1098,43) batch id : 0
[26,1] element, prob = 0.0249023    (876,29)-(918,79) batch id : 0
[27,1] element, prob = 0.0249023    (1002,69)-(1048,123) batch id : 0
[28,1] element, prob = 0.0249023    (1133,104)-(1170,163) batch id : 0
[29,1] element, prob = 0.0249023    (1056,145)-(1094,208) batch id : 0
[30,1] element, prob = 0.0249023    (861,2)-(918,63) batch id : 0
[31,1] element, prob = 0.0239258    (966,1)-(1018,45) batch id : 0
[32,1] element, prob = 0.0239258    (1098,63)-(1131,109) batch id : 0
[33,1] element, prob = 0.0239258    (1010,84)-(1092,204) batch id : 0
[34,1] element, prob = 0.0239258    (1022,139)-(1076,224) batch id : 0
[35,1] element, prob = 0.0239258    (943,34)-(1073,208) batch id : 0
[36,1] element, prob = 0.0229492    (811,3)-(851,48) batch id : 0
[37,1] element, prob = 0.0229492    (1154,27)-(1198,79) batch id : 0
[38,1] element, prob = 0.0229492    (1000,155)-(1046,226) batch id : 0
[39,1] element, prob = 0.0229492    (1086,145)-(1125,208) batch id : 0
[40,1] element, prob = 0.0229492    (958,20)-(1030,84) batch id : 0
[41,1] element, prob = 0.0229492    (934,65)-(986,155) batch id : 0
[42,1] element, prob = 0.0229492    (1045,99)-(1106,178) batch id : 0
[43,1] element, prob = 0.0229492    (983,-9)-(1141,105) batch id : 0
[44,1] element, prob = 0.0229492    (1065,-16)-(1192,124) batch id : 0
[45,1] element, prob = 0.0229492    (922,81)-(1087,304) batch id : 0
[46,1] element, prob = 0.0219727    (796,3)-(834,47) batch id : 0
[47,1] element, prob = 0.0219727    (911,4)-(953,55) batch id : 0
[48,1] element, prob = 0.0219727    (797,29)-(835,79) batch id : 0
[49,1] element, prob = 0.0219727    (821,34)-(855,81) batch id : 0
[50,1] element, prob = 0.0219727    (910,28)-(945,78) batch id : 0
[51,1] element, prob = 0.0219727    (755,77)-(790,125) batch id : 0
[52,1] element, prob = 0.0219727    (969,80)-(1008,128) batch id : 0
[53,1] element, prob = 0.0219727    (938,108)-(980,173) batch id : 0
[54,1] element, prob = 0.0219727    (1123,186)-(1161,252) batch id : 0
[55,1] element, prob = 0.0219727    (985,28)-(1056,105) batch id : 0
[56,1] element, prob = 0.0219727    (845,56)-(898,135) batch id : 0
[57,1] element, prob = 0.0219727    (1023,62)-(1080,131) batch id : 0
[58,1] element, prob = 0.0219727    (1041,40)-(1125,143) batch id : 0
[59,1] element, prob = 0.0219727    (1035,116)-(1110,242) batch id : 0
[60,1] element, prob = 0.0219727    (990,166)-(1054,269) batch id : 0
[61,1] element, prob = 0.0219727    (1078,169)-(1130,261) batch id : 0
[62,1] element, prob = 0.0219727    (971,13)-(1140,235) batch id : 0
[63,1] element, prob = 0.0219727    (980,125)-(1108,286) batch id : 0
[64,1] element, prob = 0.0209961    (940,5)-(987,55) batch id : 0
[65,1] element, prob = 0.0209961    (991,0)-(1043,41) batch id : 0
[66,1] element, prob = 0.0209961    (1126,0)-(1165,42) batch id : 0
[67,1] element, prob = 0.0209961    (998,41)-(1039,86) batch id : 0
[68,1] element, prob = 0.0209961    (1068,40)-(1103,86) batch id : 0
[69,1] element, prob = 0.0209961    (821,77)-(856,123) batch id : 0
[70,1] element, prob = 0.0209961    (913,80)-(949,128) batch id : 0
[71,1] element, prob = 0.0209961    (821,112)-(861,164) batch id : 0
[72,1] element, prob = 0.0209961    (878,113)-(919,175) batch id : 0
[73,1] element, prob = 0.0209961    (910,110)-(950,170) batch id : 0
[74,1] element, prob = 0.0209961    (1149,111)-(1189,163) batch id : 0
[75,1] element, prob = 0.0209961    (870,149)-(918,215) batch id : 0
[76,1] element, prob = 0.0209961    (1033,152)-(1070,209) batch id : 0
[77,1] element, prob = 0.0209961    (1128,141)-(1165,204) batch id : 0
[78,1] element, prob = 0.0209961    (1147,135)-(1188,199) batch id : 0
[79,1] element, prob = 0.0209961    (1064,188)-(1099,246) batch id : 0
[80,1] element, prob = 0.0209961    (1238,903)-(1279,963) batch id : 0
[81,1] element, prob = 0.0209961    (1146,19)-(1208,109) batch id : 0
[82,1] element, prob = 0.0209961    (1054,58)-(1110,126) batch id : 0
[83,1] element, prob = 0.0209961    (1111,99)-(1190,228) batch id : 0
[84,1] element, prob = 0.0209961    (1052,161)-(1105,265) batch id : 0
[85,1] element, prob = 0.0209961    (1060,145)-(1139,284) batch id : 0
[86,1] element, prob = 0.0209961    (-16,860)-(79,985) batch id : 0
[87,1] element, prob = 0.0209961    (1174,-17)-(1281,136) batch id : 0
[88,1] element, prob = 0.0209961    (1073,32)-(1198,190) batch id : 0
[89,1] element, prob = 0.0209961    (950,188)-(1072,366) batch id : 0
[90,1] element, prob = 0.0200195    (943,81)-(978,130) batch id : 0
[91,1] element, prob = 0.0200195    (968,104)-(1008,159) batch id : 0
[92,1] element, prob = 0.0200195    (930,148)-(973,211) batch id : 0
[93,1] element, prob = 0.0200195    (1092,191)-(1128,246) batch id : 0
[94,1] element, prob = 0.0200195    (1124,225)-(1161,291) batch id : 0
[95,1] element, prob = 0.0200195    (1007,-2)-(1083,55) batch id : 0
[96,1] element, prob = 0.0200195    (890,60)-(960,138) batch id : 0
[97,1] element, prob = 0.0200195    (962,65)-(1023,137) batch id : 0
[98,1] element, prob = 0.0200195    (1101,145)-(1183,284) batch id : 0
[99,1] element, prob = 0.0200195    (866,-10)-(1029,107) batch id : 0
[100,1] element, prob = 0.0200195    (1005,179)-(1111,360) batch id : 0
[101,1] element, prob = 0.0200195    (1019,342)-(1283,785) batch id : 0
[102,1] element, prob = 0.019043    (758,2)-(795,45) batch id : 0
[103,1] element, prob = 0.019043    (756,38)-(792,89) batch id : 0
[104,1] element, prob = 0.019043    (968,36)-(1011,82) batch id : 0
[105,1] element, prob = 0.019043    (729,77)-(763,127) batch id : 0
[106,1] element, prob = 0.019043    (1003,109)-(1044,169) batch id : 0
[107,1] element, prob = 0.019043    (1030,190)-(1066,242) batch id : 0
[108,1] element, prob = 0.019043    (887,83)-(971,209) batch id : 0
[109,1] element, prob = 0.019043    (995,99)-(1055,192) batch id : 0
[110,1] element, prob = 0.019043    (902,135)-(963,221) batch id : 0
[ INFO ] Image out.bmp created!
[ INFO ] Execution successful

[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
free(): invalid pointer
Aborted

For NCS 1

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
➜  samples python ./python/object_detection_sample_ssd/object_detection_sample_ssd.py -m face-detection-adas-0001.xml -d MYRIAD -i me.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Device info:
        MYRIAD
        MKLDNNPlugin version ......... 2.1
        Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
inputs number: 1
input shape: [1, 3, 384, 672]
input key: data
[ INFO ] File was added: 
[ INFO ]         me.jpg
[ WARNING ] Image me.jpg is resized from (384, 672) to (384, 672)
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Creating infer request and starting inference
[ INFO ] Processing output blobs
[0,1] element, prob = 0.871094    (420,202)-(745,608) batch id : 0 WILL BE PRINTED!
[1,1] element, prob = 0.0268555    (-54,555)-(234,1100) batch id : 0
[2,1] element, prob = 0.0209961    (21,925)-(239,1340) batch id : 0
[3,1] element, prob = 0.019043    (-18,-144)-(178,583) batch id : 0
[4,1] element, prob = 0.0180664    (636,315)-(757,470) batch id : 0
[5,1] element, prob = 0.0170898    (75,977)-(363,1321) batch id : 0
[6,1] element, prob = 0.0161133    (291,958)-(349,1123) batch id : 0
[7,1] element, prob = 0.015625    (329,962)-(391,1126) batch id : 0
[8,1] element, prob = 0.015625    (832,1121)-(937,1293) batch id : 0
[9,1] element, prob = 0.015625    (-27,670)-(136,1220) batch id : 0
[10,1] element, prob = 0.015625    (76,785)-(357,1160) batch id : 0
[11,1] element, prob = 0.015625    (204,928)-(423,1346) batch id : 0
[12,1] element, prob = 0.0146484    (694,1035)-(732,1141) batch id : 0
[13,1] element, prob = 0.0146484    (219,40)-(388,626) batch id : 0
[14,1] element, prob = 0.0136719    (658,306)-(752,405) batch id : 0
[15,1] element, prob = 0.0136719    (204,1198)-(267,1275) batch id : 0
[16,1] element, prob = 0.0136719    (322,612)-(430,793) batch id : 0
[17,1] element, prob = 0.0136719    (240,779)-(331,1046) batch id : 0
[18,1] element, prob = 0.0136719    (171,245)-(424,798) batch id : 0
[19,1] element, prob = 0.0136719    (443,205)-(802,1021) batch id : 0
[20,1] element, prob = 0.0126953    (694,511)-(731,590) batch id : 0
[21,1] element, prob = 0.0126953    (0,893)-(24,984) batch id : 0
[22,1] element, prob = 0.0126953    (272,962)-(322,1117) batch id : 0
[23,1] element, prob = 0.0126953    (268,487)-(376,688) batch id : 0
[24,1] element, prob = 0.0126953    (323,490)-(422,690) batch id : 0
[25,1] element, prob = 0.0126953    (277,608)-(384,795) batch id : 0
[26,1] element, prob = 0.0126953    (233,710)-(342,923) batch id : 0
[27,1] element, prob = 0.0126953    (256,672)-(406,957) batch id : 0
[28,1] element, prob = 0.0126953    (252,799)-(540,1176) batch id : 0
[29,1] element, prob = 0.0126953    (133,876)-(301,1420) batch id : 0
[30,1] element, prob = 0.0117188    (718,466)-(757,540) batch id : 0
[31,1] element, prob = 0.0117188    (166,527)-(206,594) batch id : 0
[32,1] element, prob = 0.0117188    (673,514)-(709,587) batch id : 0
[33,1] element, prob = 0.0117188    (-1,943)-(22,1038) batch id : 0
[34,1] element, prob = 0.0117188    (670,256)-(753,346) batch id : 0
[35,1] element, prob = 0.0117188    (705,441)-(771,555) batch id : 0
[36,1] element, prob = 0.0117188    (126,625)-(200,718) batch id : 0
[37,1] element, prob = 0.0117188    (164,673)-(251,747) batch id : 0
[38,1] element, prob = 0.0117188    (399,850)-(453,1008) batch id : 0
[39,1] element, prob = 0.0117188    (424,851)-(476,1009) batch id : 0
[40,1] element, prob = 0.0117188    (357,925)-(413,1073) batch id : 0
[41,1] element, prob = 0.0117188    (399,919)-(454,1074) batch id : 0
[42,1] element, prob = 0.0117188    (468,906)-(521,1070) batch id : 0
[43,1] element, prob = 0.0117188    (284,1016)-(357,1186) batch id : 0
[44,1] element, prob = 0.0117188    (269,379)-(374,595) batch id : 0
[45,1] element, prob = 0.0117188    (349,563)-(485,831) batch id : 0
[46,1] element, prob = 0.0117188    (252,1042)-(431,1215) batch id : 0
[47,1] element, prob = 0.0117188    (858,771)-(987,1518) batch id : 0
[48,1] element, prob = 0.0107422    (165,463)-(204,521) batch id : 0
[49,1] element, prob = 0.0107422    (185,477)-(224,537) batch id : 0
[50,1] element, prob = 0.0107422    (214,477)-(250,540) batch id : 0
[51,1] element, prob = 0.0107422    (458,478)-(488,536) batch id : 0
[52,1] element, prob = 0.0107422    (695,466)-(733,539) batch id : 0
[53,1] element, prob = 0.0107422    (460,522)-(488,590) batch id : 0
[54,1] element, prob = 0.0107422    (481,523)-(509,588) batch id : 0
[55,1] element, prob = 0.0107422    (716,512)-(755,592) batch id : 0
[56,1] element, prob = 0.0107422    (150,579)-(179,635) batch id : 0
[57,1] element, prob = 0.0107422    (171,584)-(201,641) batch id : 0
[58,1] element, prob = 0.0107422    (481,582)-(510,647) batch id : 0
[59,1] element, prob = 0.0107422    (652,565)-(689,650) batch id : 0
[60,1] element, prob = 0.0107422    (674,571)-(710,650) batch id : 0
[61,1] element, prob = 0.0107422    (694,571)-(734,648) batch id : 0
[62,1] element, prob = 0.0107422    (142,640)-(188,705) batch id : 0
[63,1] element, prob = 0.0107422    (335,627)-(390,695) batch id : 0
[64,1] element, prob = 0.0107422    (191,744)-(225,803) batch id : 0
[65,1] element, prob = 0.0107422    (370,1005)-(399,1078) batch id : 0
[66,1] element, prob = 0.0107422    (414,1006)-(443,1079) batch id : 0
[67,1] element, prob = 0.0107422    (896,1207)-(924,1280) batch id : 0
[68,1] element, prob = 0.0107422    (86,386)-(160,513) batch id : 0
[69,1] element, prob = 0.0107422    (106,477)-(214,630) batch id : 0
[70,1] element, prob = 0.0107422    (151,501)-(221,610) batch id : 0
[71,1] element, prob = 0.0107422    (428,478)-(479,615) batch id : 0
[72,1] element, prob = 0.0107422    (450,478)-(506,611) batch id : 0
[73,1] element, prob = 0.0107422    (647,498)-(728,605) batch id : 0
[74,1] element, prob = 0.0107422    (148,556)-(220,658) batch id : 0
[75,1] element, prob = 0.0107422    (425,557)-(481,671) batch id : 0
[76,1] element, prob = 0.0107422    (437,510)-(516,700) batch id : 0
[77,1] element, prob = 0.0107422    (463,554)-(527,674) batch id : 0
[78,1] element, prob = 0.0107422    (633,551)-(704,664) batch id : 0
[79,1] element, prob = 0.0107422    (280,628)-(367,711) batch id : 0
[80,1] element, prob = 0.0107422    (416,612)-(488,735) batch id : 0
[81,1] element, prob = 0.0107422    (155,706)-(260,850) batch id : 0
[82,1] element, prob = 0.0107422    (55,828)-(137,932) batch id : 0
[83,1] element, prob = 0.0107422    (451,813)-(498,960) batch id : 0
[84,1] element, prob = 0.0107422    (310,914)-(371,1076) batch id : 0
[85,1] element, prob = 0.0107422    (447,896)-(500,1065) batch id : 0
[86,1] element, prob = 0.0107422    (641,966)-(706,1121) batch id : 0
[87,1] element, prob = 0.0107422    (203,-95)-(283,182) batch id : 0
[88,1] element, prob = 0.0107422    (164,86)-(233,410) batch id : 0
[89,1] element, prob = 0.0107422    (188,498)-(307,688) batch id : 0
[90,1] element, prob = 0.0107422    (205,445)-(357,715) batch id : 0
[91,1] element, prob = 0.0107422    (-22,576)-(71,832) batch id : 0
[92,1] element, prob = 0.0107422    (230,603)-(342,791) batch id : 0
[93,1] element, prob = 0.0107422    (318,698)-(433,909) batch id : 0
[94,1] element, prob = 0.0107422    (273,784)-(390,1021) batch id : 0
[95,1] element, prob = 0.0107422    (175,740)-(444,1237) batch id : 0
[ INFO ] Image out.bmp created!
[ INFO ] Execution successful

[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
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
➜  samples python ./python/object_detection_sample_ssd/object_detection_sample_ssd.py -m face-detection-adas-0001.xml -d MYRIAD -i parents.jpg
[ INFO ] Loading Inference Engine
[ INFO ] Loading network files:
	face-detection-adas-0001.xml
	face-detection-adas-0001.bin
[ INFO ] Device info:
        MYRIAD
        MKLDNNPlugin version ......... 2.1
        Build ........... 2020.3.0-3467-15f2c61a-releases/2020/3
inputs number: 1
input shape: [1, 3, 384, 672]
input key: data
[ INFO ] File was added: 
[ INFO ]         parents.jpg
[ WARNING ] Image parents.jpg is resized from (384, 672) to (384, 672)
[ INFO ] Preparing input blobs
[ INFO ] Batch size is 1
[ INFO ] Preparing output blobs
[ INFO ] Loading model to the device
[ INFO ] Creating infer request and starting inference
[ INFO ] Processing output blobs
[0,1] element, prob = 1.0    (826,366)-(1026,644) batch id : 0 WILL BE PRINTED!
[1,1] element, prob = 0.996582    (539,173)-(693,429) batch id : 0 WILL BE PRINTED!
[2,1] element, prob = 0.552734    (1094,47)-(1135,95) batch id : 0 WILL BE PRINTED!
[3,1] element, prob = 0.22168    (848,22)-(886,70) batch id : 0
[4,1] element, prob = 0.0537109    (8,784)-(151,956) batch id : 0
[5,1] element, prob = 0.0395508    (1033,78)-(1070,122) batch id : 0
[6,1] element, prob = 0.034668    (1123,75)-(1158,122) batch id : 0
[7,1] element, prob = 0.0322266    (1086,14)-(1138,85) batch id : 0
[8,1] element, prob = 0.03125    (1091,110)-(1130,171) batch id : 0
[9,1] element, prob = 0.0302734    (1108,43)-(1150,93) batch id : 0
[10,1] element, prob = 0.0292969    (1151,83)-(1190,131) batch id : 0
[11,1] element, prob = 0.0273438    (1091,6)-(1133,55) batch id : 0
[12,1] element, prob = 0.0273438    (1064,73)-(1098,121) batch id : 0
[13,1] element, prob = 0.0268555    (1030,2)-(1073,42) batch id : 0
[14,1] element, prob = 0.0268555    (1058,113)-(1096,168) batch id : 0
[15,1] element, prob = 0.0268555    (1047,-3)-(1110,59) batch id : 0
[16,1] element, prob = 0.0268555    (1142,75)-(1211,165) batch id : 0
[17,1] element, prob = 0.0268555    (849,14)-(1030,236) batch id : 0
[18,1] element, prob = 0.0258789    (941,22)-(981,74) batch id : 0
[19,1] element, prob = 0.0258789    (1030,36)-(1074,89) batch id : 0
[20,1] element, prob = 0.0258789    (876,73)-(915,118) batch id : 0
[21,1] element, prob = 0.0258789    (845,112)-(886,167) batch id : 0
[22,1] element, prob = 0.0258789    (1036,110)-(1076,171) batch id : 0
[23,1] element, prob = 0.0258789    (1113,-1)-(1175,59) batch id : 0
[24,1] element, prob = 0.0258789    (1003,41)-(1094,151) batch id : 0
[25,1] element, prob = 0.0249023    (1059,0)-(1098,43) batch id : 0
[26,1] element, prob = 0.0249023    (1002,69)-(1048,123) batch id : 0
[27,1] element, prob = 0.0249023    (1133,105)-(1170,164) batch id : 0
[28,1] element, prob = 0.0239258    (966,1)-(1018,45) batch id : 0
[29,1] element, prob = 0.0239258    (876,29)-(918,79) batch id : 0
[30,1] element, prob = 0.0239258    (1056,145)-(1094,208) batch id : 0
[31,1] element, prob = 0.0239258    (861,2)-(918,63) batch id : 0
[32,1] element, prob = 0.0239258    (1010,84)-(1092,204) batch id : 0
[33,1] element, prob = 0.0239258    (1022,139)-(1076,224) batch id : 0
[34,1] element, prob = 0.0239258    (942,35)-(1072,208) batch id : 0
[35,1] element, prob = 0.0229492    (811,3)-(851,48) batch id : 0
[36,1] element, prob = 0.0229492    (1098,63)-(1131,109) batch id : 0
[37,1] element, prob = 0.0229492    (983,-9)-(1141,105) batch id : 0
[38,1] element, prob = 0.0229492    (921,81)-(1087,304) batch id : 0
[39,1] element, prob = 0.0219727    (796,3)-(835,47) batch id : 0
[40,1] element, prob = 0.0219727    (911,4)-(953,55) batch id : 0
[41,1] element, prob = 0.0219727    (797,29)-(835,79) batch id : 0
[42,1] element, prob = 0.0219727    (1154,27)-(1198,79) batch id : 0
[43,1] element, prob = 0.0219727    (969,80)-(1008,128) batch id : 0
[44,1] element, prob = 0.0219727    (1000,155)-(1046,226) batch id : 0
[45,1] element, prob = 0.0219727    (1086,145)-(1125,208) batch id : 0
[46,1] element, prob = 0.0219727    (1123,186)-(1161,252) batch id : 0
[47,1] element, prob = 0.0219727    (958,20)-(1030,84) batch id : 0
[48,1] element, prob = 0.0219727    (985,28)-(1056,105) batch id : 0
[49,1] element, prob = 0.0219727    (845,56)-(898,134) batch id : 0
[50,1] element, prob = 0.0219727    (934,65)-(986,155) batch id : 0
[51,1] element, prob = 0.0219727    (1046,99)-(1107,178) batch id : 0
[52,1] element, prob = 0.0219727    (1035,116)-(1110,241) batch id : 0
[53,1] element, prob = 0.0219727    (990,166)-(1054,269) batch id : 0
[54,1] element, prob = 0.0219727    (1065,-16)-(1192,124) batch id : 0
[55,1] element, prob = 0.0219727    (971,13)-(1140,235) batch id : 0
[56,1] element, prob = 0.0219727    (980,125)-(1108,286) batch id : 0
[57,1] element, prob = 0.0209961    (991,0)-(1043,41) batch id : 0
[58,1] element, prob = 0.0209961    (1126,0)-(1165,42) batch id : 0
[59,1] element, prob = 0.0209961    (821,35)-(856,79) batch id : 0
[60,1] element, prob = 0.0209961    (910,28)-(945,78) batch id : 0
[61,1] element, prob = 0.0209961    (997,41)-(1038,86) batch id : 0
[62,1] element, prob = 0.0209961    (1068,40)-(1103,86) batch id : 0
[63,1] element, prob = 0.0209961    (754,77)-(789,125) batch id : 0
[64,1] element, prob = 0.0209961    (819,77)-(855,123) batch id : 0
[65,1] element, prob = 0.0209961    (913,79)-(949,128) batch id : 0
[66,1] element, prob = 0.0209961    (821,112)-(861,164) batch id : 0
[67,1] element, prob = 0.0209961    (938,108)-(980,173) batch id : 0
[68,1] element, prob = 0.0209961    (1149,112)-(1189,162) batch id : 0
[69,1] element, prob = 0.0209961    (870,149)-(918,215) batch id : 0
[70,1] element, prob = 0.0209961    (1033,152)-(1070,209) batch id : 0
[71,1] element, prob = 0.0209961    (1064,188)-(1099,246) batch id : 0
[72,1] element, prob = 0.0209961    (1023,63)-(1081,131) batch id : 0
[73,1] element, prob = 0.0209961    (1054,58)-(1110,126) batch id : 0
[74,1] element, prob = 0.0209961    (1041,40)-(1125,143) batch id : 0
[75,1] element, prob = 0.0209961    (1111,99)-(1190,228) batch id : 0
[76,1] element, prob = 0.0209961    (1078,169)-(1130,261) batch id : 0
[77,1] element, prob = 0.0209961    (950,189)-(1072,367) batch id : 0
[78,1] element, prob = 0.0200195    (940,6)-(987,56) batch id : 0
[79,1] element, prob = 0.0200195    (943,81)-(978,130) batch id : 0
[80,1] element, prob = 0.0200195    (878,114)-(919,175) batch id : 0
[81,1] element, prob = 0.0200195    (910,110)-(950,170) batch id : 0
[82,1] element, prob = 0.0200195    (968,104)-(1008,159) batch id : 0
[83,1] element, prob = 0.0200195    (1128,141)-(1165,204) batch id : 0
[84,1] element, prob = 0.0200195    (1147,135)-(1188,199) batch id : 0
[85,1] element, prob = 0.0200195    (1238,903)-(1279,963) batch id : 0
[86,1] element, prob = 0.0200195    (1007,-2)-(1083,55) batch id : 0
[87,1] element, prob = 0.0200195    (1052,161)-(1105,266) batch id : 0
[88,1] element, prob = 0.0200195    (1061,145)-(1140,284) batch id : 0
[89,1] element, prob = 0.0200195    (-16,860)-(79,985) batch id : 0
[90,1] element, prob = 0.0200195    (866,-10)-(1029,107) batch id : 0
[91,1] element, prob = 0.0200195    (1173,-17)-(1280,136) batch id : 0
[92,1] element, prob = 0.0200195    (1072,32)-(1197,189) batch id : 0
[93,1] element, prob = 0.0200195    (1005,179)-(1111,359) batch id : 0
[94,1] element, prob = 0.019043    (758,2)-(795,45) batch id : 0
[95,1] element, prob = 0.019043    (968,35)-(1011,82) batch id : 0
[96,1] element, prob = 0.019043    (1003,109)-(1044,169) batch id : 0
[97,1] element, prob = 0.019043    (930,148)-(973,211) batch id : 0
[98,1] element, prob = 0.019043    (1092,191)-(1128,246) batch id : 0
[99,1] element, prob = 0.019043    (1123,225)-(1161,291) batch id : 0
[100,1] element, prob = 0.019043    (1146,19)-(1208,109) batch id : 0
[101,1] element, prob = 0.019043    (897,59)-(958,143) batch id : 0
[102,1] element, prob = 0.019043    (962,65)-(1023,137) batch id : 0
[103,1] element, prob = 0.019043    (886,83)-(971,209) batch id : 0
[104,1] element, prob = 0.019043    (995,99)-(1055,192) batch id : 0
[105,1] element, prob = 0.019043    (902,135)-(963,221) batch id : 0
[106,1] element, prob = 0.019043    (1101,145)-(1183,284) batch id : 0
[107,1] element, prob = 0.019043    (821,-17)-(946,129) batch id : 0
[108,1] element, prob = 0.019043    (940,-16)-(1076,120) batch id : 0
[109,1] element, prob = 0.019043    (819,34)-(945,205) batch id : 0
[110,1] element, prob = 0.019043    (1019,342)-(1283,785) batch id : 0
[111,1] element, prob = 0.0180664    (1147,7)-(1193,54) batch id : 0
[112,1] element, prob = 0.0180664    (756,38)-(792,89) batch id : 0
[ INFO ] Image out.bmp created!
[ INFO ] Execution successful

[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
corrupted double-linked list
Aborted

3.3.4 Results

Parents At First Starbucks Parents Face Detection By Object Detection SSD NCS 1 Parents Face Detection By Object Detection SSD NCS 2
Parents At First Starbucks Parents At First Starbucks - Face Detection By Object Detection SSD NCS 1 Parents At First Starbucks - Face Detection By Object Detection SSD NCS 2
Me Me Face Detection By Object Detection SSD NCS 1 Me Face Detection By OpenCV DNN NCS 2
Me Me - Face Detection By Object Detection SSD NCS 1 Me - Face Detection By Object Detection SSD NCS 2

3.4 Model Optimization

In this section, we are going to test out another OpenVINO example: Image Classification C++ Sample Async. After reading this blog, it wouldn’t be hard for us to notice the MOST important thing we’re missing here is the model file alexnet_fp32.xml. Let’s just keep it in mind for now.

And, let’s review a bit about our previous example: Object Detection – we downloaded face-detection-adas-0001 model from online and use it directly. So, questions:

  • Are we able to download alexnet_fp32.xml from online this time again?
  • Where can we download a whole bunch of open source models?

3.4.1 Open Model Zoo

It wouldn’t be hard for us to google out OpenVINO™ Toolkit - Open Model Zoo repository, under which model face-detection-adas-0001 is just sitting there. However, face-detection-adas-0001.xml and face-detection-adas-0001.bin are missing.

Let’s checkout open_model_zoo and put it under folder /opt/intel.

1
2
3
4
5
6
7
8
➜  intel pwd
/opt/intel
➜  intel ll
total 12K
drwxr-xr-x 8 pi   pi   4.0K Apr  7 06:37 dldt
drwxr-xr-x 9 pi   pi   4.0K Apr  7 14:04 l_openvino_toolkit_runtime_raspbian_p_2020.1.023
drwxr-xr-x 7 pi   pi   4.0K Apr  7 14:04 open_model_zoo
lrwxrwxrwx 1 root root   48 Apr  7 06:29 openvino -> l_openvino_toolkit_runtime_raspbian_p_2020.1.023

Then, let's enter folder face-detection-adas-0001 under open_model_zoo and take a look:

1
2
3
4
➜  face-detection-adas-0001 git:(master) pwd
/opt/intel/open_model_zoo/models/intel/face-detection-adas-0001
➜  face-detection-adas-0001 git:(master) ls
description  face-detection-adas-0001.prototxt  model.yml

It seems that each folder under intel and public contains the detailed info of each model. For instance, file intel/face-detection-adas-0001/model.yml contains all the info about model face-detection-adas-0001. However, what we really need are a .xml file and a .bin file. In the following, we are going to generate such 2 files, which are optimized specifically for movidius by following Intel OpenVINO toolkit issue 798441.

3.4.2 Download Caffe Model Files

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
➜  downloader git:(master) pwd
/opt/intel/open_model_zoo/tools/downloader
➜  downloader git:(master) ls
caffe2_to_onnx.py  converter.py   info_dumper.py      README.md               requirements.in          tests
common.py          downloader.py  pytorch_to_onnx.py  requirements-caffe2.in  requirements-pytorch.in
➜  downloader git:(master) python downloader.py --name alexnet
################|| Downloading models ||################

========== Downloading /opt/intel/open_model_zoo/tools/downloader/public/alexnet/alexnet.prototxt
... 100%, 3 KB, 10485 KB/s, 0 seconds passed

========== Downloading /opt/intel/open_model_zoo/tools/downloader/public/alexnet/alexnet.caffemodel
... 100%, 238146 KB, 5428 KB/s, 43 seconds passed

################|| Post-processing ||################

========== Replacing text in /opt/intel/open_model_zoo/tools/downloader/public/alexnet/alexnet.prototxt
➜  downloader git:(master) ✗ ll public/alexnet/
total 233M
-rw-r--r-- 1 pi pi 233M Apr  9 14:24 alexnet.caffemodel
-rw-r--r-- 1 pi pi 3.6K Apr  9 14:24 alexnet.prototxt
-rw-r--r-- 1 pi pi 3.6K Apr  9 14:23 alexnet.prototxt.orig

Clearly, three files including a large model file alexnet.caffemodel has been downloaded.

3.4.3 Model Optimization

Now, are are going to optimize the downloaded caffe model and make it feedable to OpenVINO.

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
➜  model-optimizer git:(2020) ✗ pwd
/opt/intel/dldt/model-optimizer
➜  model-optimizer git:(2020) ✗ python mo_caffe.py --generate_deprecated_IR_V7 --input_model /opt/intel/open_model_zoo/tools/downloader/public/alexnet/alexnet.caffemodel
[ WARNING ]  Use of deprecated cli option --generate_deprecated_IR_V7 detected. Option use in the following releases will be fatal. 
Model Optimizer arguments:
Common parameters:
	- Path to the Input Model: 	/opt/intel/open_model_zoo/tools/downloader/public/alexnet/alexnet.caffemodel
	- Path for generated IR: 	/opt/intel/dldt/model-optimizer/.
	- IR output name: 	alexnet
	- Log level: 	ERROR
	- Batch: 	Not specified, inherited from the model
	- Input layers: 	Not specified, inherited from the model
	- Output layers: 	Not specified, inherited from the model
	- Input shapes: 	Not specified, inherited from the model
	- Mean values: 	Not specified
	- Scale values: 	Not specified
	- Scale factor: 	Not specified
	- Precision of IR: 	FP32
	- Enable fusing: 	True
	- Enable grouped convolutions fusing: 	True
	- Move mean values to preprocess section: 	False
	- Reverse input channels: 	False
Caffe specific parameters:
	- Path to Python Caffe* parser generated from caffe.proto: 	mo/front/caffe/proto
	- Enable resnet optimization: 	True
	- Path to the Input prototxt: 	/opt/intel/open_model_zoo/tools/downloader/public/alexnet/alexnet.prototxt
	- Path to CustomLayersMapping.xml: 	Default
	- Path to a mean file: 	Not specified
	- Offsets for a mean file: 	Not specified
Model Optimizer version: 	unknown version
Please expect that Model Optimizer conversion might be slow. You are currently using Python protobuf library implementation. 
However you can use the C++ protobuf implementation that is supplied with the OpenVINO toolkitor build protobuf library from sources. 
Navigate to "install_prerequisites" folder and run: python -m easy_install protobuf-3.5.1-py($your_python_version)-win-amd64.egg 
set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=cpp


 For more information please refer to Model Optimizer FAQ (https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_Model_Optimizer_FAQ.html), question #80. 

[ SUCCESS ] Generated IR version 7 model.
[ SUCCESS ] XML file: /opt/intel/dldt/model-optimizer/./alexnet.xml
[ SUCCESS ] BIN file: /opt/intel/dldt/model-optimizer/./alexnet.bin
[ SUCCESS ] Total execution time: 615.36 seconds. 
[ SUCCESS ] Memory consumed: 1982 MB.

And, let's take a look at what's generated under folder /opt/intel/dldt/model-optimizer.

1
2
3
4
5
6
7
8
9
10
➜  model-optimizer git:(2020) ✗ pwd
/opt/intel/dldt/model-optimizer
➜  model-optimizer git:(2020) ✗ ls
alexnet.bin      extensions             mo_caffe.py  mo_onnx.py  README.md               requirements_kaldi.txt  requirements_tf.txt
alexnet.mapping  install_prerequisites  mo_kaldi.py  mo.py       requirements_caffe.txt  requirements_mxnet.txt  requirements.txt
alexnet.xml      mo                     mo_mxnet.py  mo_tf.py    requirements_dev.txt    requirements_onnx.txt   tf_call_ie_layer
➜  model-optimizer git:(2020) ✗ ll alexnet*
-rw-r--r-- 1 pi pi 233M Apr  9 15:26 alexnet.bin
-rw-r--r-- 1 pi pi 2.6K Apr  9 15:26 alexnet.mapping
-rw-r--r-- 1 pi pi  25K Apr  9 15:26 alexnet.xml

Note: You may meet the following ERRORs during model optimization.

1
2
3
4
[ ERROR ]  
Detected not satisfied dependencies:
	networkx: installed: 2.4, required: 2.4
	protobuf: installed: 3.11.3, required: 3.6.1

Clearly, for networkx, the ERROR message is a kind of ridiculous. Anyway, if you meet the above 2 errors, please DOWNGRADE your packages as follows:

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
➜  ~  pip install protobuf==3.6.1 --user
Looking in indexes: https://pypi.org/simple, https://www.piwheels.org/simple
Collecting protobuf==3.6.1
  Downloading https://files.pythonhosted.org/packages/77/78/a7f1ce761e2c738e209857175cd4f90a8562d1bde32868a8cd5290d58926/protobuf-3.6.1-py2.py3-none-any.whl (390kB)
     |████████████████████████████████| 399kB 1.4MB/s 
Requirement already satisfied: six>=1.9 in /home/pi/.local/lib/python3.7/site-packages (from protobuf==3.6.1) (1.13.0)
Requirement already satisfied: setuptools in /home/pi/.local/lib/python3.7/site-packages (from protobuf==3.6.1) (44.0.0)
Installing collected packages: protobuf
  Found existing installation: protobuf 3.11.3
    Uninstalling protobuf-3.11.3:
      Successfully uninstalled protobuf-3.11.3
Successfully installed protobuf-3.6.1
➜  ~  pip install networkx==2.3 --user
Looking in indexes: https://pypi.org/simple, https://www.piwheels.org/simple
Collecting networkx==2.3
  Downloading https://files.pythonhosted.org/packages/85/08/f20aef11d4c343b557e5de6b9548761811eb16e438cee3d32b1c66c8566b/networkx-2.3.zip (1.7MB)
     |████████████████████████████████| 1.8MB 2.3MB/s 
Requirement already satisfied: decorator>=4.3.0 in /home/pi/.local/lib/python3.7/site-packages (from networkx==2.3) (4.4.1)
Building wheels for collected packages: networkx
  Building wheel for networkx (setup.py) ... done
  Created wheel for networkx: filename=networkx-2.3-py2.py3-none-any.whl size=1556408 sha256=9964dc8e3b41e97f0228afb2c1c6ca40a5dfc07dc1a2574b92717114f3d5e533
  Stored in directory: /home/pi/.cache/pip/wheels/de/63/64/3699be2a9d0ccdb37c7f16329acf3863fd76eda58c39c737af
Successfully built networkx
Installing collected packages: networkx
  Found existing installation: networkx 2.4
    Uninstalling networkx-2.4:
      Successfully uninstalled networkx-2.4
Successfully installed networkx-2.3

3.5 Image Classification

3.5.1 C

For NCS 2

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
➜  samples ./c/build/armv7l/Release/hello_classification_c /opt/intel/dldt/model-optimizer/alexnet.xml me.jpg HETERO:MYRIAD

Top 10 results:

Image me.jpg

classid probability
------- -----------
838       0.074951
978       0.072632
977       0.072083
975       0.060730
903       0.052948
638       0.040588
976       0.040161
433       0.037842
112       0.032471
639       0.029724
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
➜  samples ./c/build/armv7l/Release/hello_classification_c /opt/intel/dldt/model-optimizer/alexnet.xml parents.jpg HETERO:MYRIAD

Top 10 results:

Image parents.jpg

classid probability
------- -----------
707       0.072144
577       0.024185
704       0.022980
955       0.020844
813       0.020432
910       0.016937
515       0.016235
605       0.013985
918       0.013351
523       0.013031

For NCS 1

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
➜  samples ./c/build/armv7l/Release/hello_classification_c /opt/intel/dldt/model-optimizer/alexnet.xml me.jpg HETERO:MYRIAD

Top 10 results:

Image me.jpg

classid probability
------- -----------
978       0.082336
977       0.081665
838       0.081055
975       0.061188
903       0.054840
638       0.046387
433       0.040131
976       0.034607
112       0.032349
639       0.031494
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
➜  samples ./c/build/armv7l/Release/hello_classification_c /opt/intel/dldt/model-optimizer/alexnet.xml parents.jpg HETERO:MYRIAD

Top 10 results:

Image parents.jpg

classid probability
------- -----------
707       0.059418
577       0.026993
813       0.021347
704       0.020218
910       0.019669
515       0.018204
955       0.018051
523       0.013527
808       0.012856
918       0.012802

3.5.2 C++

For NCS 2

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
➜  samples ./cpp/build/armv7l/Release/hello_classification /opt/intel/dldt/model-optimizer/alexnet.xml me.jpg HETERO:MYRIAD

Top 10 results:

Image me.jpg

classid probability
------- -----------
838     0.0749512  
978     0.0726318  
977     0.0720825  
975     0.0607300  
903     0.0529480  
638     0.0405884  
976     0.0401611  
433     0.0378418  
112     0.0324707  
639     0.0297241  

This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
➜  samples ./cpp/build/armv7l/Release/hello_classification /opt/intel/dldt/model-optimizer/alexnet.xml parents.jpg HETERO:MYRIAD

Top 10 results:

Image parents.jpg

classid probability
------- -----------
707     0.0721436  
577     0.0241852  
704     0.0229797  
955     0.0208435  
813     0.0204315  
910     0.0169373  
515     0.0162354  
605     0.0139847  
918     0.0133514  
523     0.0130310  

This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool

For NCS 1

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
➜  samples ./cpp/build/armv7l/Release/hello_classification /opt/intel/dldt/model-optimizer/alexnet.xml me.jpg HETERO:MYRIAD               

Top 10 results:

Image me.jpg

classid probability
------- -----------
978     0.0823364  
977     0.0816650  
838     0.0810547  
975     0.0611877  
903     0.0548401  
638     0.0463867  
433     0.0401306  
976     0.0346069  
112     0.0323486  
639     0.0314941  

This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
➜  samples ./cpp/build/armv7l/Release/hello_classification /opt/intel/dldt/model-optimizer/alexnet.xml parents.jpg HETERO:MYRIAD

Top 10 results:

Image parents.jpg

classid probability
------- -----------
707     0.0594177  
577     0.0269928  
813     0.0213470  
704     0.0202179  
910     0.0196686  
515     0.0182037  
955     0.0180511  
523     0.0135269  
808     0.0128555  
918     0.0128021  

This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool

3.5.3 Python

For NCS 2

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
➜  samples python ./python/classification_sample/classification_sample.py -m /opt/intel/dldt/model-optimizer/alexnet.xml -i me.jpg -d HETERO:MYRIAD -nt 10
[ INFO ] Creating Inference Engine
[ INFO ] Loading network files:
	/opt/intel/dldt/model-optimizer/alexnet.xml
	/opt/intel/dldt/model-optimizer/alexnet.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image me.jpg is resized from (1280, 924) to (227, 227)
[ INFO ] Batch size is 1
[ INFO ] Loading model to the plugin
[ INFO ] Starting inference in synchronous mode
[ INFO ] Processing output blob
[ INFO ] Top 10 results: 
Image me.jpg

classid probability
------- -----------
  838    0.0751343
  978    0.0728149
  977    0.0722656
  975    0.0606079
  903    0.0533142
  638    0.0413513
  976    0.0391541
  433    0.0375061
  112    0.0320740
  639    0.0298767


[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
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
➜  samples python ./python/classification_sample/classification_sample.py -m /opt/intel/dldt/model-optimizer/alexnet.xml -i parents.jpg -d HETERO:MYRIAD -nt 10
[ INFO ] Creating Inference Engine
[ INFO ] Loading network files:
	/opt/intel/dldt/model-optimizer/alexnet.xml
	/opt/intel/dldt/model-optimizer/alexnet.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image parents.jpg is resized from (960, 1280) to (227, 227)
[ INFO ] Batch size is 1
[ INFO ] Loading model to the plugin
[ INFO ] Starting inference in synchronous mode
[ INFO ] Processing output blob
[ INFO ] Top 10 results: 
Image parents.jpg

classid probability
------- -----------
  707    0.0712891
  577    0.0237122
  704    0.0227814
  955    0.0209045
  813    0.0205231
  910    0.0169983
  515    0.0162354
  605    0.0140991
  918    0.0133438
  523    0.0129318


[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool

For NCS 1

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
➜  samples python ./python/classification_sample/classification_sample.py -m /opt/intel/dldt/model-optimizer/alexnet.xml -i me.jpg -d HETERO:MYRIAD -nt 10
[ INFO ] Creating Inference Engine
[ INFO ] Loading network files:
	/opt/intel/dldt/model-optimizer/alexnet.xml
	/opt/intel/dldt/model-optimizer/alexnet.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image me.jpg is resized from (1280, 924) to (227, 227)
[ INFO ] Batch size is 1
[ INFO ] Loading model to the plugin
[ INFO ] Starting inference in synchronous mode
[ INFO ] Processing output blob
[ INFO ] Top 10 results: 
Image me.jpg

classid probability
------- -----------
  978    0.0818481
  977    0.0818481
  838    0.0812378
  975    0.0610657
  903    0.0553894
  638    0.0469971
  433    0.0398865
  976    0.0338440
  112    0.0317993
  639    0.0317993


[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool
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
➜  samples python ./python/classification_sample/classification_sample.py -m /opt/intel/dldt/model-optimizer/alexnet.xml -i parents.jpg -d HETERO:MYRIAD -nt 10
[ INFO ] Creating Inference Engine
[ INFO ] Loading network files:
	/opt/intel/dldt/model-optimizer/alexnet.xml
	/opt/intel/dldt/model-optimizer/alexnet.bin
[ INFO ] Preparing input blobs
[ WARNING ] Image parents.jpg is resized from (960, 1280) to (227, 227)
[ INFO ] Batch size is 1
[ INFO ] Loading model to the plugin
[ INFO ] Starting inference in synchronous mode
[ INFO ] Processing output blob
[ INFO ] Top 10 results: 
Image parents.jpg

classid probability
------- -----------
  707    0.0590515
  577    0.0267334
  813    0.0212250
  704    0.0199432
  910    0.0197906
  955    0.0183716
  515    0.0182190
  523    0.0135040
  808    0.0129242
  918    0.0126648

3.5.4 Results

Classification Result for me.jpg

Result from NCS 2 Result from NCS 1
838: 'sunscreen, sunblock, sun blocker', 978: 'seashore, coast, seacoast, sea-coast',
978: 'seashore, coast, seacoast, sea-coast', 977: 'sandbar, sand bar',
977: 'sandbar, sand bar', 838: 'sunscreen, sunblock, sun blocker',
975: 'lakeside, lakeshore', 975: 'lakeside, lakeshore',
903: 'wig', 903: 'wig',
638: 'maillot', 638: 'maillot',
976: 'promontory, headland, head, foreland', 433: 'bathing cap, swimming cap',
433: 'bathing cap, swimming cap', 976: 'promontory, headland, head, foreland',
112: 'conch', 112: 'conch',
639: 'maillot, tank suit', 639: 'maillot, tank suit',

Classification Result for parents.jpg

Result from NCS 2 Result from NCS 1
707: 'pay-phone, pay-station', 707: 'pay-phone, pay-station',
577: 'gong, tam-tam', 577: 'gong, tam-tam',
704: 'parking meter', 813: 'spatula',
955: 'jackfruit, jak, jack', 704: 'parking meter',
813: 'spatula', 910: 'wooden spoon',
910: 'wooden spoon', 515: 'cowboy hat, ten-gallon hat',
515: 'cowboy hat, ten-gallon hat', 955: 'jackfruit, jak, jack',
605: 'iPod', 523: 'crutch',
918: 'crossword puzzle, crossword', 808: 'sombrero',
523: 'crutch', 918: 'crossword puzzle, crossword',

classid table:

4. OpenCV DNN with OpenVINO’s Inference Engine

For the example openvino_fd_myriad.py given on Install OpenVINO™ toolkit for Raspbian* OS, the TOUGH thing is how to build OpenCV with OpenVINO’s Inference Engine. NEVER forget to export export InferenceEngine_DIR="/opt/intel/openvino/inference_engine/share" and enable WITH_INF_ENGINE ON to have OpenCV rebuilt. Before moving forward, my modified version of openvino_fd_myriad.py is provided as follows:

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
import sys, argparse
import cv2

parser = argparse.ArgumentParser(description='OpenVINO Face Detection on MYRIAD')
parser.add_argument('-x', '--XML', help='Required: model XML file')
parser.add_argument('-b', '--BIN', help='Required: model BIN file')
parser.add_argument('-i', '--input', help='Required: input image')
parser.add_argument('-o', '--output', help='Not Required: output resultant image, by default: out.jpg')


args = parser.parse_args()
xmlfile = args.XML
binfile = args.BIN
inputimage = args.input
outputimage = args.output

if not outputimage:
    outputimage = "out.jpg"

if not xmlfile:
    print(f'openvino_fd_myriad {args.xmlfile}')

if not binfile:
    print(f'openvino_fd_myriad {args.binfile}')
    
if not inputimage:
    print(f'openvino_fd_myriad {args.inputimage}')


       
# Load the model.
net = cv2.dnn.readNet(xmlfile, binfile)
# Specify target device.
net.setPreferableTarget(cv2.dnn.DNN_TARGET_MYRIAD)
# Read an image.
frame = cv2.imread(inputimage)
if frame is None:
    raise Exception('Image not found!')
# Prepare input blob and perform an inference.
print("Frame Size: {}".format(frame.shape))
blob = cv2.dnn.blobFromImage(frame, scalefactor=1.0, size=(224, 224), swapRB=True, ddepth=cv2.CV_32F)
net.setInput(blob)
print("First Blob: {}".format(blob.shape))
out = net.forward()
count = 0
# Draw detected faces on the frame.
for detection in out.reshape(-1, 7):
    confidence = float(detection[2])
    xmin = int(detection[3] * frame.shape[1])
    ymin = int(detection[4] * frame.shape[0])
    xmax = int(detection[5] * frame.shape[1])
    ymax = int(detection[6] * frame.shape[0])
    if confidence > 0.8:
        cv2.rectangle(frame, (xmin, ymin), (xmax, ymax), color=(0, 255, 0))
        count += 1
print(count)
# Save the frame to an image file.
cv2.imwrite(outputimage, frame)

4.1 Intel64

On my laptop, of course, we are building OpenCV for architecture Intel64, with NVidia GPU + CUDA support, for DNN in OpenCV requires either CUDA or OpenCL.

My test result of openvino_fd_myriad.py shows the performance of adopted model face-detection-adas-0001 is NOT as good as expected.

For NCS 2

1
2
3
4
5
6
7
8
➜  OpenVINO_Examples python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.bin -i ./parents.jpg
Frame Size: (960, 1280, 3)
First Blob: (1, 3, 224, 224)
2
➜  OpenVINO_Examples python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.bin -i ./me.jpg
Frame Size: (1280, 924, 3)
First Blob: (1, 3, 224, 224)
1

For NCS 1

1
2
3
4
5
6
7
8
➜  OpenVINO_Examples python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.bin -i ./parents.jpg
Frame Size: (960, 1280, 3)
First Blob: (1, 3, 224, 224)
2
➜  OpenVINO_Examples python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/cpp/build_18.04/intel64/Release/face-detection-adas-0001.bin -i ./me.jpg
Frame Size: (1280, 924, 3)
First Blob: (1, 3, 224, 224)
1

Results

Parents At First Starbucks Parents Face Detection By OpenCV DNN NCS 1 Parents Face Detection By OpenCV DNN NCS 2
Parents At First Starbucks Parents At First Starbucks - Face Detection By NCS 1 Parents At First Starbucks - Face Detection By NCS 2
Me Me Face Detection By OpenCV DNN NCS 1 Me Face Detection By OpenCV DNN NCS 2
Me Me - Face Detection By OpenCV DNN NCS 1 Me - Face Detection By OpenCV DNN NCS 2

4.2 armv7l

For NCS 2

1
2
3
4
5
6
7
8
➜  Programs python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.bin -i ./parents.jpg
Frame Size: (960, 1280, 3)
First Blob: (1, 3, 224, 224)
2
➜  Programs python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.bin -i ./me.jpg
Frame Size: (1280, 924, 3)
First Blob: (1, 3, 224, 224)
1

For NCS 1

1
2
3
4
5
6
7
8
➜  Programs python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.bin -i ./parents.jpg
Frame Size: (960, 1280, 3)
First Blob: (1, 3, 224, 224)
2
➜  Programs python ./openvino_fd_myriad.py -x /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.xml -b /opt/intel/openvino/inference_engine/samples/face-detection-adas-0001.bin -i ./me.jpg
Frame Size: (1280, 924, 3)
First Blob: (1, 3, 224, 224)
1

Results

Parents At First Starbucks Parents Face Detection By OpenCV DNN NCS 1 Parents Face Detection By OpenCV DNN NCS 2
Parents At First Starbucks Parents At First Starbucks - Face Detection By NCS 1 Parents At First Starbucks - Face Detection By NCS 2
Me Me Face Detection By OpenCV DNN NCS 1 Me Face Detection By OpenCV DNN NCS 2
Me Me - Face Detection By OpenCV DNN NCS 1 Me - Face Detection By OpenCV DNN NCS 2

5. My Built Raspbian ISO With OpenCV4 + OpenVINO

Finally, you are welcome to try my built image rpi4-raspbian20200213-opencv4.3-openvino2020.1.023-ncsdk2.10.01.01.img, which is about 9.5G and composed of:

Everything has ALREADY been updated and built successfully, ONLY EXCEPT object_detection_sample_ssd. In addition, rpi4-raspbian20200213-opencv4.3-openvino2020.1.023-ncsdk2.10.01.01.img ALSO works properly on my Raspberry Pi 3 Model B Version 1.2 manufactured in 2015.