I'm Priyanka from Intel.

In this video, we give you a summary

of all the videos in this series.

We also introduce some next steps, more advanced examples,

and dev kits to easily get started with your computer

visions solutions.

[MUSIC PLAYING]

In the first video, we introduced you

to this computer vision series with Intel video solutions

for computer vision and deep learning applications.

In the second video, we introduced the OpenVINO toolkit

to accelerate computer vision application development

across Intel platforms.

We also talk about different components of the OpenVINO

toolkit to help optimize your deep learning inference.

Then we look at the model optimizer,

an important component of the deep learning deployment

toolkit, to do model conversion.

We discuss the model conversion techniques

and presented an example to convert a pre-trained model

to an intermediate format using model optimizer.

After that, in the fourth video, we deep

dive into the inference engine to run optimized inference

on different Intel platforms using a unified API.

We also look at a simple example to demonstrate inference engine

API usage and run the application on the CPU

and the integrated GPU.

Then, in the fifth video, we talk about the hetero plugin

from the inference engine to support hardware heterogeneity.

We also demonstrate running the computer vision application

on the Intel Movidius Compute Stick.

In the sixth video, we discuss optimization techniques

using OpenVINO toolkit to get better performance

for your computer vision application.

Finally, we discuss advanced video analytics

using OpenVINO toolkit and pre-trained models

included in the release package to expedite computer vision

application development.

To wrap up the series, I will walk you

through additional resources to use and help ramp

up on OpenVINO toolkit.

Along with core samples in the release package,

we also have included tutorials for building end

to end IoT reference solutions, addressing specific business

use cases.

For example, the store traffic monitor

reference implementation uses multiple video streams

that count people inside and outside of a facility,

and also counts product inventory.

You can check other reference implementations,

such as face access control, intruder detector, and people

counter from the links provided.

IEI and Aaeon introduced two developer kits

to help developers get started quickly

with their vision application development using Intel tools.

These kits ship with Intel System Studio and OpenVINO

pre-installed.

This makes them vision ready out of the box,

giving you the chance to try some of the code samples

and demos with just a few clicks.

To learn about tools, core samples, reference

implementations, and developer kits,

you can follow the links provided.

Thanks for watching the computer vision with Intel smart

video tools series.

[MUSIC PLAYING]