How to run OMZ demos

You could verify IE ARM CPU plugin by running OMZ demos. If ARM CPU plugin is built using pre-configured Dokerfiles with enabled WITH_OMZ_DEMO then you may find compiled C++ OMZ demos at <INSTALL_DIR>/extras/open_model_zoo/demos/Release. Otherwise, you need to build OMZ demos on ARM platform.

Build OMZ demos

1. Run the setupvars.sh script to set all necessary environment variables:

source <INSTALL_DIR>/setupvars.sh

2. Clone OMZ repository:

git clone --recurse-submodules --single-branch --branch=master https://github.com/openvinotoolkit/open_model_zoo.git

3. Go to open_model_zoo\demos directory and run the build_demos.sh script:

./build_demos.sh

Once the build is completed, you can find sample binaries in ~/omz_demos_build/aarch64/Release

Run OMZ demos

Before running compiled binary files, make sure your application can find the Inference Engine and OpenCV libraries. Run the setupvars.sh script to set all necessary environment variables:

source <INSTALL_DIR>/setupvars.sh

NOTE: The OpenVINO environment variables are removed when you close the shell. As an option, you can permanently set the environment variables by adding the command source <INSTALL_DIR>/setupvars.sh to the file ~/.bashrc.

You are ready to run OMZ demos. Each demo application supports -h flag which prints help message and usage examples. To run OMZ demos you may use videos and photos from here and here.

NOTE: Python OMZ demo applications could be run as well as C++ applications on Linux only. Python applications depend on scipy module which can't be installed on Apple M1 so far: scipy#13409.