Installing OpenVINO™ packages¶
During the ros-humble-openvino-node
package installation you will be asked
if you want to install the OpenVINO™ IR formatted models.
Some of the OpenVINO™ based tutorials in this SDK rely on the models that are
provided during the installation of the ros-humble-openvino-node
.
It is important to answer this question with “yes”.
Configuring ros-humble-openvino-node
------------------------------------
The Intel OpenVINO™ toolkit provides a set of public models and Intel's pre-trained models in OpenVINO™ IR format (*.xml + *.bin) that you can use for learning and demo purposes or for developing deep learning on ROS2 node.
Do you want to download & convert OpenVINO™ IR formatted models [default=no]? [yes/no] yes
Note
In case you are facing issues running the tutorials because you failed to install the OpenVINO models:
Cleanup OpenVINO by following the steps in OpenVINO™ Re-Installation and Troubleshooting.
Reinstall OpenVINO following OpenVINO™ Installation Steps.
OpenVINO™ Installation Steps¶
Make sure OpenVINO APT repository is configured
wget -O- https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS.PUB | gpg --dearmor | sudo tee /usr/share/keyrings/openvino-archive-keyring.gpg echo "deb [signed-by=/usr/share/keyrings/openvino-archive-keyring.gpg] https://apt.repos.intel.com/openvino/2023 ubuntu22 main" | sudo tee /etc/apt/sources.list.d/intel-openvino-2023.list
Install get
debconf
utilitiesClear any previous installation configurations
Set/force the config option to True before installation of the package
This step can’t be automated since it would violate licensing terms by not prompting, but if the user manually does the config it is probably fine.
OpenVINO™ Re-Installation and Troubleshooting¶
To reinstall OpenVINO™ or cleanup after a failed installation: