All language subtitles for 3. About this course - Copy

So while in the open Zeevi in Python.

Well Python is becoming more and more popular and that's because it's one of the easiest languages for

beginners.

It allows us to focus on building complex computer vision apps without being bogged down by the intricacies

of the language itself.

And I'm looking at you C++.

However Python is still extremely powerful especially for science and machine learning applications

which are an essential part of the can be division world.

Finally it allows us to store images and non-payers which allows us to do some very powerful operations

quite easily.

So let's take a quick look at exactly what you'll be learning on this course so fiercely given the current

state of can be division.

I try to give you an excellent foundation that exposes you to all key areas of computer vision.

We start off by doing the basics where we get into some simple image manipulations and segmentation.

We then implement some basic object detection followed by feature detection and call and people detection

as well.

We don't take a look at this analysis and filters.

After that we go into some basic machine learning in computer vision and then we get into some more

motion analysis and object tracking.

I've also included a short mini project based on competition of photography and then we wrap up the

course where I give you some advice and resources on how to become an expert and can be division.

I also show you some of the latest research areas and also give you some very cool startup ideas that

involve computer.

Computer vision and best of all in discourse.

You get to implement almost 50 different computer vision exercises and implement 12 very fun many projects.

So it should come as no surprise that this is a very practical course where we're going to spend more

than half of our time coding.

However before we dive into it could always teach the theory first before hand unless I'm using the

code to actually teach at that topic as well.

And the is always explained line by line except in cases where it becomes a bit redundant.

Or the theory at hand is a bit too complex for this group of discourse.

So I may have mentioned 12 mini projects before.

So what exactly are these mini projects.

Let's take a look.

So here they are in all their glory all 12 many projects on one slide so fiercely.

You're getting to make an awesome Live sketch of yourself using a webcam.

We don't get to implement is simple ship matching project followed by an app that actually comes a number

of circles and ellipses in an image.

We don't move on to.

We're finding Waldo projec followed by a simple object detection project.

You don't get to implement fi's pedestrian encored detection.

After that you get to implement a very cool life swapping up here where you can play a Donald Trump

or Kim Kardashian's or anyone else's face in real time.

And then we implement a simple human detection up after which you get to make a basic machine learning

app that actually understands handwritten digits.

This is then followed by a face recognition app and you don't get to implement a simple ball trucking

up and I know this isn't a ball it's actually a clock.

I couldn't find my ball but that's OK.

And lastly we get to do a simple photo restoration app we can remove this line from his photo right

here.

So you're definitely getting a lot of practical experience making computer vision applications so I

really hope you enjoy doing these projects.

The requirements for the scores are actually pretty low.

Basic programming would actually be very helpful as well as exposure to non-pay.

However it's not needed as I actually go through the code line by line.

Secondly a high school level math would actually be very good to have to appreciate some of the high

level concepts that we're implementing.

And also you need to have a webcam to implement a lot of for many projects as well as some of the example

code.

Now we're going to install Pitre an open C.V right after in the next section.

However I'll just point out that I used the Anaconda package solution and that allows me to use Pitre

notebooks which are excellent for teaching since it since it allows us to use and or uncovered in court

blocks.

Now there is some unfortunate news regarding the latest version of Open C-v which is true point 1.

It unfortunately no longer support some important functions such as swift and Souf which I use for object

detection.

So I would recommend you install 2.4 one tree instead.

However that said there are some object tracking techniques that aren't supported in 2.4 and tree.

So depending on what's your priority you can choose which version you would like to install accordingly.

So who exactly is this course for.

Well I've designed this course to suit a wide number of people starting from beginners who just have

an interest in computer vision or even software developers and engineers looking to strengthen their

job skills as well as college and university students looking to get a head start in the computer vision

projects and research also startup founders who wish to use law as some sort of computer vision component

to their companies.

And finally hobbyists who just want to build some fun computer vision project using a Raspberry Pi perhaps.

So let's begin our exciting journey into the world of computer vision using open C-v in Python.