Original subtitles

1

Hello and welcome back to the course on computer vision in today's tutorial We'll talk about the idea

2

behind Gad's.

3

Guns were invented by a good fellow or to be entirely correct by Dr good fellow who currently works

4

at openly as a researcher and he's only 31 years old.

5

He actually created Gannes even earlier so when he was in in his late 20s a couple of years ago he created

6

Gans and the story goes that he finished off the idea of guns and when he was at a bar none of that's

7

true or not but you can definitely check out more of that online.

8

But it's a crazy story anyway.

9

How such inside such a young age has brought something to the world that is really drastically changing

10

this whole field of neural networks and computer vision.

11

And so what are Gad's the creation grandstands for generative adversarial network.

12

And we're going to talk more about what Gannes are and how they're shocked at how they're trained what

13

they're used for in the further trolls in this section.

14

But I just want to talk about the philosophy.

15

Why were they created in the first place.

16

Well the reason why Gannes were constructed is because neural networks are generally good at predicting

17

things or classifying things and solving problems.

18

But we wanted to create or I guess to say that researchers AI researchers wanted to create a type of

19

neural network that we can create for itself and that's why they came up with this idea of generative

20

adversarial networks.

21

So Gans can actually generate images they can create images of all kinds of photos of things that never

22

actually ever existed before.

23

So they learn about the world they learn about our objects that we have that we use or animals or anything.

24

And then they can create new versions of those objects that actually never walk this planet or never

25

these objects never existed were never built by humans but they can create images of those objects so

26

they can actually if you see you can think about it as they can have an imagination.

27

And the way that Gannes work is they have two components.

28

They have a generator and a discriminator.

29

And these two components are constantly in touch.

30

The generator generates images and the discriminator then assesses those images and tells the generator

31

whether or not those images are likely to be similar or are similar to what it has actually seen in

32

the real world that already exists.

33

So it's got these two components at work in tandem.

34

And when you're training this is the interesting part.

35

This is like a very philosophical part really that will definitely go into the detail about how the

36

generator works how it works how they connect and so on.

37

But what we need to understand is the philosophical concept that when you're creating a degenerative

38

adversarial network when you're training up you are bringing these two up together from scratch so it's

39

not like you have.

40

And this is the part where I needed to get my head around when I was learning about them is that you're

41

not just creating is it's like you have a discriminator that already knows everything about the world.

42

And then it is this generator is generating things and the discriminator can't right away tell the generator

43

if that is you know that looks like a real object or not.

44

No the fact is that they start from scratch from zero and then they learn together.

45

So the generator generate some images for instance you're generating images of tables and then the discriminator

46

will look at the images that the generated generated and will look at some images of real tables to

47

compare the two and it will learn for itself what is a table what isn't a table because it all know

48

but by default it will know what the generator generates actually doesn't exist in it.

49

It has that knowledge and therefore it will know what is a table what isn't a table and then it will

50

also give feedback to the generator saying hey your tables are not really tables because of this and

51

this and this.

52

I could tell right away.

53

So they were kind of playing a game there.

54

Think of it as are against each other but they're learning together.

55

So the general is always trying to create images which look like real tables to fool the discriminator

56

to trick the discriminator in believing that those are real tables whereas the real discrimination always

57

has actual real to images of actual real tables to compare to in order to learn better.

58

And then also give feedback to the generator.

59

And in that sense and we'll see exactly the mechanics of this for it in that sense they learn together

60

they grow together and eventually this whole network as a whole it learns how to better and better and

61

better create these tables that are indistinguishable from the real world tables and zone even though

62

they're working against each other fighting against each other as a whole the network is getting better

63

and better and better and more robust.

64

So that's the philosophy behind Gannes that they are designed to create objects or images to be to be

65

exactly and be entirely accurate images of objects animals or even humans based on what we have in the

66

real world.

67

So you can think of it as a general adversarial network who's actually learning about our world and

68

then creating objects or images of objects which actually are very similar to what we have in our world

69

but never actually existed.

70

So yeah that's us.

71

I hope that sounds exciting.

72

We're going to delve into that in the upcoming tutorials.

73

And for now if you'd like some additional reading there's a great blog post by Chan chan SORN.

74

So torn.

75

I hope I pronounced that right.

76

It's on Hakkar noon and you can check it out there.

77

I've got a link.

78

Basically it will talk you through the intuition behind Ganns.

79

Well we're going to discuss all of this in upcoming tutorials but in case you wanted to prepare in case

80

you wanted to do some pre-reading so that your you know you have an additional point of view then you

81

can check out some extra insights over there.

82

But again we will discuss all of this in upcoming tutorials.

83

And I can't wait to see you there.

84

Until next time.

85

Enjoy computer vision.