Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6404 Discussions

deploy my application on rasp os with intel movidus

momo93
Novice
761 Views

Hello 

how we can deploy our applications from developpement environnement such wind10 or linux to Raspberry Pi, I already installed openvino on rasp os 32 and samples works but with the other application from wind10 or linux doesnt 

thanks 

0 Kudos
4 Replies
Iffa_Intel
Moderator
717 Views

Greetings,


You may refer here for the official OpenVINO Deployment Manager guide:

https://docs.openvinotoolkit.org/latest/openvino_docs_install_guides_deployment_manager_tool.html


The Deployment Manager of Intel® Distribution of OpenVINO™ creates a deployment package by assembling the model, IR files, your application, and associated dependencies into a runtime package for your target device.


Sincerely,

Iffa


0 Kudos
momo93
Novice
713 Views

hello 

Thank your answer,  for Deployment Manage tool the  The operating system on the target host must be the same as the development system on which we are creating the package. in case we are working in linux or even wind10. we are in 64 bit environnement , and in the other hand  raspberry pi os is an 32 bits. that's is the real problem. 

0 Kudos
Iffa_Intel
Moderator
698 Views

Greetings,


The cross-platform for OpenVINO deployment is supported.

The documentation I had provided before does have the instruction for both windows and Linux.


To generate the openvino_deployment_package (both .zip and .tar) on development machine,

 

cd <INSTALL_DIR>/deployment_tools/tools/deployment_manager/

python3 /deployment_manager.py

 

choose which devices to run the inference (CPU, GPU, or NCS2), then it will generate the .zip or .tar on the same folder.

 

  1. copy the openvino_deployment_package.tar.gz/.zip to the target host.
  2. unzip the file
  3. execute the environment setup
  4. run inference sample code.

 

Please note that, if the target machine is Linux, the deploment_manager must be generated in Linux development machines and the same goes for Windows environments

 

For RaspberryPi, we do not have much documentation that we can generate the deployment manager from Linux to RasberryPi since both have a different environment.

 

If you are trying out for Raspberry pi, you can try to generate the deployment manager in a Linux environment and ensure to select NCS2 because RaspberryPi doesn't work on CPU.


Sincerely,

Iffa



0 Kudos
Iffa_Intel
Moderator
659 Views

Greetings,


Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question


Sincerely,

Iffa


0 Kudos
Reply