Community
cancel
Showing results for 
Search instead for 
Did you mean: 
SashaBelykh
Beginner
89 Views

Up Core Plus + Up Vision Plus X (with 3 Myriad X VPUs) software issues

Hello.

Summary: I'm looking for a guide from people who already started inference on all 3 VPUs embedded to Up Vison Plus X board.

I have received AAEON SBC Up Core Plus with the brand new carrier/expansion board known as Up Vision Plus X, which bears 3 Myriad X VPUs and promises very interesting possibilities in a pretty small design.

Unfortunately, how it is usual for the new products, there is a lack of documentation and developer experience with this board.

I have faced some issues, but first, let me explain system hardware and done steps:

1. System hardware:

1.1. UPC-PLUSX5Q-A20-0464 a.k.a. Up Core Plus with  RAM: 4GB; Flash: 64GB; Intel® Atom™ x5 E3940 SoC

1.2. UPCP-CR-VPX3-A10-001 a.k.a. Up Vision Plus X: Vision accelerator with 3 x  Movidius Myriad X VPU

2. Steps followed at the moment:

- Ubuntu 18.04 has been installed.

- Intel OpenVINO distribution 2019R1.1 has been installed, following steps from the official guide. Namely, Core Components and Steps for Intel® Movidius™ Neural Compute Stick and Intel® Neural Compute Stick 2 have been successfully performed.

 

Now, the issue.

I have successfully executed demo applications on CPU and NCS2 but failed to run it on "embedded" VPUs, which are suited on the carrier board. And even successfully executed case of use NCS2 reproduces some error messages.

Here are the logs from the command line:

Running /opt/intel/openvino_2019.1.144/deployment_tools/demo$ ./demo_security_barrier_camera.sh  -d MYRIAD


… some scripted preparations …

Application start:

###################################################

Run Inference Engine security_barrier_camera demo

Run ./security_barrier_camera_demo -d MYRIAD -d_va MYRIAD -d_lpr MYRIAD -i /opt/intel/openvino_2019.1.144/deployment_tools/demo/car_1.bmp -m /home/sasha/openvino_models/ir/FP16/Security/object_detection/barrier/0106/dldt/vehicle-license-plate-detection-barrier-0106-fp16.xml -m_va /home/sasha/openvino_models/ir/FP16/Security/object_attributes/vehicle/resnet10_update_1/dldt/vehicle-attributes-recognition-barrier-0039-fp16.xml -m_lpr /home/sasha/openvino_models/ir/FP16/Security/optical_character_recognition/license_plate/dldt/license-plate-recognition-barrier-0001-fp16.xml 

[ INFO ] InferenceEngine: 
        API version ............ 1.6
        Build .................. custom_releases/2019/R1.1_28dfbfdd28954c4dfd2f94403dd8dfc1f411038b
[ INFO ] Parsing input parameters
[ INFO ] Capturing video streams from the video files or loading images
[ INFO ] Files were added: 1
[ INFO ]     /opt/intel/openvino_2019.1.144/deployment_tools/demo/car_1.bmp
[ INFO ] Number of input image files: 1
[ INFO ] Number of input video files: 0
[ INFO ] Number of input channels: 1
[ INFO ] Display resolution: 1920x1080
[ INFO ] Loading plugin MYRIAD

        API version ............ 1.6
        Build .................. 23780
        Description ....... myriadPlugin
[ INFO ] Loading network files for VehicleDetection
[ INFO ] Batch size is forced to  1
[ INFO ] Checking Vehicle Detection inputs
[ INFO ] Checking Vehicle Detection outputs
[ INFO ] Loading Vehicle Detection model to the MYRIAD plugin
<strong>E: [ncAPI] [    293174] ncDeviceOpen:800        Failed connection to device (1.5.1-ma2480) with error 5</strong>
[ ERROR ] Can not init USB device: NC_DEVICE_NOT_FOUND
Error on or near line 228; exiting with status 1

 

Same script with inserted NCS2:

###################################################

Run Inference Engine security_barrier_camera demo

Run ./security_barrier_camera_demo -d MYRIAD -d_va MYRIAD -d_lpr MYRIAD -i /opt/intel/openvino_2019.1.144/deployment_tools/demo/car_1.bmp -m /home/sasha/openvino_models/ir/FP16/Security/object_detection/barrier/0106/dldt/vehicle-license-plate-detection-barrier-0106-fp16.xml -m_va /home/sasha/openvino_models/ir/FP16/Security/object_attributes/vehicle/resnet10_update_1/dldt/vehicle-attributes-recognition-barrier-0039-fp16.xml -m_lpr /home/sasha/openvino_models/ir/FP16/Security/optical_character_recognition/license_plate/dldt/license-plate-recognition-barrier-0001-fp16.xml 

[ INFO ] InferenceEngine: 
        API version ............ 1.6
        Build .................. custom_releases/2019/R1.1_28dfbfdd28954c4dfd2f94403dd8dfc1f411038b
[ INFO ] Parsing input parameters
[ INFO ] Capturing video streams from the video files or loading images
[ INFO ] Files were added: 1
[ INFO ]     /opt/intel/openvino_2019.1.144/deployment_tools/demo/car_1.bmp
[ INFO ] Number of input image files: 1
[ INFO ] Number of input video files: 0
[ INFO ] Number of input channels: 1
[ INFO ] Display resolution: 1920x1080
[ INFO ] Loading plugin MYRIAD

        API version ............ 1.6
        Build .................. 23780
        Description ....... myriadPlugin
[ INFO ] Loading network files for VehicleDetection
[ INFO ] Batch size is forced to  1
[ INFO ] Checking Vehicle Detection inputs
[ INFO ] Checking Vehicle Detection outputs
[ INFO ] Loading Vehicle Detection model to the MYRIAD plugin
[ INFO ] Loading network files for VehicleAttribs
[ INFO ] Batch size is forced to 1 for Vehicle Attribs
[ INFO ] Checking VehicleAttribs inputs
[ INFO ] Checking Vehicle Attribs outputs
[ INFO ] Loading Vehicle Attribs model to the MYRIAD plugin
E: [ncAPI] [    624758] ncDeviceOpen:672        Failed to find suitable device, rc: X_LINK_DEVICE_NOT_FOUND
[ INFO ] Loading network files for Licence Plate Recognition (LPR)
[ INFO ] Batch size is forced to  1 for LPR Network
[ INFO ] Checking LPR Network inputs
[ INFO ] Checking LPR Network outputs
[ INFO ] Loading LPR model to the MYRIAD plugin
E: [ncAPI] [    626812] ncDeviceOpen:672        Failed to find suitable device, rc: X_LINK_DEVICE_NOT_FOUND
[ INFO ] Start inference 
To close the application, press 'CTRL+C' or any key with focus on the output window

Average inference time: 48.7759 ms (20.5019 fps)

Average vehicle detection time: 31.711 ms (31.5348 fps)

Average vehicle attribs time: 5.1753 ms (193.226 fps)

Average lpr time: 11.0977 ms (90.1085 fps)

Total execution time: 3434.22

[ INFO ] Execution successful


###################################################

Demo completed successfully.

In order to compare and be sure that NCS2 actually did the things, same script started without -d MYRIAD option (on Intel CPU) and it shows the following performance:

Average inference time: 48.098 ms (20.7909 fps)

Average vehicle detection time: 21.1481 ms (47.2857 fps)

Average vehicle attribs time: 9.6757 ms (103.352 fps)

Average lpr time: 22.2226 ms (44.9993 fps)

Total execution time: 4611.07

[ INFO ] Execution successful


###################################################

Demo completed successfully.

Despite the strange similarity of Average inference time, average times of inference stages look realy accelerated by NCS2. But it still does not help with embedded VPUs.

It would be greatly appreciated if someone who already received the same board can provide guidance to setup and tune such system and perform inference on all three embedded VPUs.

Best regard

Belykh Aleksandr.

0 Kudos
3 Replies
Shubha_R_Intel
Employee
89 Views

Dear Belykh, Aleksandr

It's hard to say where the problem is.

Here are a few suggestions:

  1. Double check USB rules, ensure that all additional installation steps are performed
  2. If you have HDDL service running in the background (it is there if you had run any sample with a HDDL plugin even once) -- so If you did – then the sure way is to reboot, or OpenVino 2019 R2 release documentation will come with instructions on how to correctly stop it (just killing processes is not enough). Unfortunately R2 is not released yet (but will be soon).
  3. Reboot (don't shutdown) the hardware. 

Please try these steps and report back with your progress.

And thanks for your patience !

Shubha

 

 

SashaBelykh
Beginner
89 Views

Dear Shubha, thank you for your answer!

Nice to see you again!

I have checked the second set and it works! Most likely, the root cause of the mentioned issue was in this blue label.

Engineering Sample Best regards

Aleksandr.

Shubha_R_Intel
Employee
89 Views

Dear Belykh, Aleksandr,

I'm very happy that everything works for you now ! Thanks for reporting back to the community !

Shubha

Reply