Intel® OpenVINO™ Docker Environment

What is OpenVINO™?

The OpenVINO™ Toolkit’s name comes from “Open Visual Inferencing and Neural Network Optimization.” It is primarily focused around optimizing neural network inference and is open source. OpenVino doesn’t support training.

Intel developed this tool to helps support fast inference across Intel® CPUs, GPUs, FPGAs, and Neural Compute Stick with a standard API.

OpenVINO™ can take models built with multiple different frameworks, like TensorFlow or Caffe, and use its Model Optimizer to optimize for inference.

This optimized model used with the Inference Engine that helps speed inference on the related hardware. It also has a wide variety of Pre-Trained Models already put through Model Optimizer.

In OpenVINO™, Pre-Trained Models refer specifically to the Model Zoo, in which the Free Model Set contains pre-trained models already converted using the Model Optimizer. These models can be used directly with the Inference Engine.

OpenVINO™ Docker Image:
Prerequisites:
1. Docker setup
2. Git
Build a Docker image:

git clone https://github.com/nullbyte91/Intel-edge-AI-foundation-udacity.git
cd Intel-edge-AI-foundation-udacity/docker/
docker build — rm — build-arg proxy=$http_proxy — rm — tag openvino:v1 .

Run a container:

docker run -v /etc/localtime:/etc/localtime:ro — rm -it -e http_proxy -e https_proxy -e ftp_proxy -v `pwd`:/work openvino:v1 bash

Running GUI apps with Docker:
We need to share the Host X11 display server with a docker container to see the image window when we run an application.

docker run -v /etc/localtime:/etc/localtime:ro -it -e DISPLAY=$DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix -e http_proxy -e https_proxy -e ftp_proxy -v `pwd`:/work openvino:v1 bash

Mount Camera:
To perform both real-time (webcam) and video processing, We need to mount a camera with the container.

docker run -v /etc/localtime:/etc/localtime:ro — rm -it — device /dev/video0 -e http_proxy -e https_proxy -e ftp_proxy -v `pwd`:/work openvino:v1 bash

Test your environment:
Finally, we are going to test our docker environment with one demo application.

cd /opt/intel/openvino/deployment_tools/demo/
bash demo_security_barrier_camera.sh