Create a Yocto Image with Intel® Distribution of OpenVINO™ toolkit

This document provides instructions for creating a Yocto image with Intel® Distribution of OpenVINO™ toolkit.

System Requirements

Use the Yocto Project official documentation to set up and configure your host machine to be compatible with BitBake.

Step 1: Set Up Environment

Set Up Git Repositories

The following Git repositories are required to build a Yocto image:

Clone these Git repositories to your host machine:

git clone https://git.yoctoproject.org/git/poky --branch honister
git clone https://git.yoctoproject.org/git/meta-intel --branch honister
git clone https://git.openembedded.org/meta-openembedded --branch honister
git clone https://github.com/kraj/meta-clang.git --branch honister

Set up BitBake Layers

source poky/oe-init-build-env
bitbake-layers add-layer ../meta-intel
bitbake-layers add-layer ../meta-openembedded/meta-oe
bitbake-layers add-layer ../meta-openembedded/meta-python
bitbake-layers add-layer ../meta-clang

Set up BitBake Configurations

Include extra configuration in conf/local.conf in your build directory as required.

# Build with SSE4.2, AVX2 etc. extensions
MACHINE = "intel-skylake-64"

# Enable clDNN GPU plugin when needed.
# This requires meta-clang and meta-oe layers to be included in bblayers.conf
# and is not enabled by default.
PACKAGECONFIG_append_pn-openvino-inference-engine = " opencl"

# Enable building inference engine python API.
# This requires meta-python layer to be included in bblayers.conf.
PACKAGECONFIG_append_pn-openvino-inference-engine = " python3"

# This adds inference engine related libraries in the target image.
CORE_IMAGE_EXTRA_INSTALL_append = " openvino-inference-engine"

# This adds inference engine samples in the target image.
CORE_IMAGE_EXTRA_INSTALL_append = " openvino-inference-engine-samples"

# Include inference engine python API package in the target image.
CORE_IMAGE_EXTRA_INSTALL_append = " openvino-inference-engine-python3"

# Enable MYRIAD plugin
CORE_IMAGE_EXTRA_INSTALL_append = " openvino-inference-engine-vpu-firmware"

# Include model optimizer in the target image.
CORE_IMAGE_EXTRA_INSTALL_append = " openvino-model-optimizer"

Step 2: Build a Yocto Image with OpenVINO Packages

Run BitBake to build your image with OpenVINO packages. To build the minimal image, for example, run:

bitbake core-image-minimal

Step 3: Verify the Yocto Image with OpenVINO Packages

Verify that OpenVINO packages were built successfully. Run the following command:

oe-pkgdata-util list-pkgs | grep openvino

If the image was built successfully, it will return the list of packages as below:

openvino-inference-engine
openvino-inference-engine-dbg
openvino-inference-engine-dev
openvino-inference-engine-python3
openvino-inference-engine-samples
openvino-inference-engine-src
openvino-inference-engine-staticdev
openvino-inference-engine-vpu-firmware
openvino-model-optimizer
openvino-model-optimizer-dbg
openvino-model-optimizer-dev