I'm Priyanka Bagade.

I'm a developer evangelist at Intel.

I train developers on the latest Intel IOT technologies

through workshops, hackathons, and training videos.

In this video series, we explore Intel's smart video tools

for computer vision applications.

[MUSIC PLAYING]

With the internet of things, more and more systems

are getting connected to the internet, where

we can analyze the sensor data to monitor and control

the systems.

Cameras are one of the most important sensors.

The video application opportunities are endless.

For example, advanced medical research, personalized health

care, smart transportation, smart cities, manufacturing,

retail, or supply chain management.

These industries rely on video for critical insights

and competitive growth.

Considering such a large amount of data

is generated by these systems, deep learning

seems to be a more robust solution

for video analytics over traditional computer vision.

Deep learning can help to extract meaningful information

from the available data.

For example, when processing images or videos,

it can detect objects, faces, and emotions

from the millions of pixels of the image.

Intel has been working on solutions for video

to understand developers needs.

We then address those needs using different platforms,

such as smart cameras, video gateways, NVRs,

and data centers.

Additionally, we offer tools, such as the OpenVINO and Media

SDK, to get accelerate video analytics at edge.

In this video series, we go into details of the OpenVINO Toolkit

to do optimized inference at the edge for computer vision

applications.

In the second video of this series,

we introduce you to OpenVINO Toolkit

to do video analytics at the edge.

We talk about components of the OpenVINO Toolkit

and the new programming model to deploy application

on a range of silicon by Intel.

The third video dives into the model optimizer,

which is one of the main components of OpenVINO

Toolkit for model conversion.

After that, in the fourth video, we

cover the inference engine, which provides a unified API

to run the application on different hardware types.

Then in the fifth video, we cover

hardware heterogeneity plugin and how

to run the application on different hardware

types, such as CPU, GPU, Movidius Compute Stick,

and FPGA using the inference engine API.

In the sixth video, we talk about optimization techniques.

In the seventh video, we discuss advanced video analytics

using the OpenVINO Toolkit.

In the final video, we provide a summary of entire series.

We also give you some next steps and more advanced examples.

At the end of the series, you should

be able to write an optimized inference

application at the edge using the OpenVINO Toolkit.

Thanks for watching.

Watch the next video in this series for an introduction

to OpenVINO Toolkit.

[INTEL THEME]