Sie sind auf Seite 1von 3

Getting started with OpenVINO

- Purpose
+ Declaring what we can do with OpenVINO as a developer.
+ Guiding a new guy to be familiar with OpenVINO through doing some demonstrations.
- Reference
[1] Bring Intelligence to the Edge with Intel® Movidius™ Neural Compute Stick
https://software.intel.com/en-us/videos/bring-intelligence-to-the-edge-with-intel-movidius-neural-
compute-stick
[2] Install Intel® Distribution of OpenVINO™ toolkit for Linux* with FPGA Support
https://docs.openvinotoolkit.org/latest/_docs_install_guides_installing_openvino_linux_fpga.html
[3] inference-tutorials-generic
https://github.com/intel-iot-devkit/inference-tutorials-generic/tree/openvino_toolkit_2019_r1_0
[4] Face Detection Tutorial
https://github.com/intel-iot-devkit/inference-tutorials-
generic/tree/openvino_toolkit_2019_r1_0/face_detection_tutorial
[5] OpenVINO™ Toolkit - Open Model Zoo repository
https://github.com/opencv/open_model_zoo/tree/master
[6] Inference Engine Demos
https://github.com/opencv/open_model_zoo/tree/master/demos
[7] Smart Classroom C++ Demo
https://github.com/opencv/open_model_zoo/tree/master/demos/smart_classroom_demo
[8] OpenVINO™ toolkit -- English
https://www.youtube.com/watch?v=kY9nZbX1DWM&list=PLDKCjIU5YH6jMzcTV5_cxX9aPHs
borbXQ
[9] Inference Engine Samples
https://docs.openvinotoolkit.org/latest/_docs_IE_DG_Samples_Overview.html
1. What are we focusing on?
1.1. As a part of Applications developers

Image 1. Data scientist or Applications developer [1]

In a real project, there are some guys:


- Applications developers
- Data scientists
What we are focusing now is a part of Applications developers. We will get a model from Data
scientists and produce output for the other Applications developers.
1.2. What is OpenVINO? [8]
OpenVINO is a toolkit that you can use to develop and deploy vision-oriented solutions on Intel
platforms. OpenVINO stands for Open Visual Inferencing and Neural Network Optimization. [8]
OpenVINO is focused on Deep Learning inference, not training.
This tool is very useful when your mind arises a question that how do you use a model to do the
detection or classification on new image or video.
Image 2.
Inference
flow with
OpenVI
NO [4]

OpenVI
NO
toolkit
includes
the
followin
g
components:
Imag
e 3.
Com
pone
nts of
Open
VIN
O [8]
-
Mode
l
Opti
mizer
-
Infere
nce
Engin
e
- FPGA RunTime Environment
- Bitstreams

2. Installation OpenVINO and getting demo data


2.1. Installation OpenVINO
Following the guide in [2].
Note: chose the right guide that is relative with the Intel® Distribution of OpenVINO™ toolkit
package.
Image 4. Corresponding guide version of OpenVINO package
2.2. Getting demo data
Following the tutorial in [4]
This demo will show you how to a problem in the real world, using Deep Learning with Intel’s
tools.

3. How to run a demo


OpenVINO is coming with many code samples and pre-trained models that are easy to use and get
started.
Following the introduction in [9] to install the samples demos.

WHAT NEXT?

Das könnte Ihnen auch gefallen