Torsten Reil: Using biology to make better animation

33,203 views ใƒป 2008-07-08

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

ืžืชืจื’ื: Alon Dotan ืžื‘ืงืจ: Sigal Tifferet
00:15
I'm going to talk about a technology that we're developing at Oxford now,
0
15160
4000
ืื ื™ ืื“ื‘ืจ ืขืœ ื˜ื›ื ื•ืœื•ื’ื™ื” ืฉืื ื—ื ื• ืžืคืชื—ื™ื ื‘ืื•ืงืกืคื•ืจื“ ื‘ื™ืžื™ื ืืœื•
00:19
that we think is going to change the way that
1
19160
3000
ืฉืื ื—ื ื• ื—ื•ืฉื‘ื™ื ืฉืชืฉื ื” ืืช ื”ื“ืจืš ื‘ื”
00:22
computer games and Hollywood movies are being made.
2
22160
3000
ืžืฉื—ืงื™ ืžื—ืฉื‘ ื•ืกืจื˜ื™ ื”ื•ืœื™ื•ื•ื“ ื ืขืฉื™ื.
00:26
That technology is simulating humans.
3
26160
3000
ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ื”ื–ื• ืžื“ืžื” ืื ืฉื™ื
00:29
It's simulated humans with a simulated body
4
29160
3000
ื”ื™ื ืžื“ืžื” ืื ืฉื™ื ืขื ื’ื•ืฃ ืžื“ื•ืžื”
00:32
and a simulated nervous system to control that body.
5
32160
3000
ื•ืžืขืจื›ืช ืขืฆื‘ื™ื ืžื“ื•ืžื” ืฉืฉื•ืœื˜ืช ื‘ื’ื•ืฃ ื–ื”
00:36
Now, before I talk more about that technology,
6
36160
3000
ืขื›ืฉื™ื• ืœืคื ื™ ืฉืืžืฉื™ืš ืœื“ื‘ืจ ืขืœ ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ื”ื–ื•
00:39
let's have a quick look at what human characters look like
7
39160
3000
ื‘ื•ืื• ื ืจืื” ืื™ืš ื“ืžื•ื™ื•ืช ืื ื•ืฉื™ื•ืช ื ืจืื•ืช
00:42
at the moment in computer games.
8
42160
3000
ื›ื™ื•ื ื‘ืžืฉื—ืงื™ ืžื—ืฉื‘
00:45
This is a clip from a game called "Grand Theft Auto 3."
9
45160
3000
ื–ื”ื• ืงืœื™ืค ืžืžืฉื—ืง ืžื—ืฉื‘ ืฉื ืงืจื "ื’ื ื™ื‘ืช ื”ืžื›ื•ื ื™ื•ืช ื”ื’ื“ื•ืœื” 3"
00:48
We already saw that briefly yesterday.
10
48160
2000
ื™ืฆื ืœื ื• ืœืจืื•ืช ืืช ื–ื” ื‘ืงืฆืจื” ืืชืžื•ืœ
00:50
And what you can see is -- it is actually a very good game.
11
50160
3000
ื•ืžื” ืฉืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ื–ื”, ื”ืืžืช ืฉื–ื”ื• ืžืฉื—ืง ื˜ื•ื‘ ืžืื“
00:53
It's one of the most successful games of all time.
12
53160
3000
ื–ื” ืื—ื“ ืžืžืฉื—ืงื™ ื”ืžื—ืฉื‘ ื”ื˜ื•ื‘ื™ื ื‘ื›ืœ ื”ื–ืžื ื™ื
00:56
But what you'll see is that all the animations in this game are very repetitive.
13
56160
4000
ืื‘ืœ ืžื” ืฉืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ื–ื” ืฉื”ืื ื™ืžืฆื™ื•ืช ื‘ืžืฉื—ืง ื—ื•ื–ืจื•ืช ืขืœ ืขืฆืžืŸ
01:00
They pretty much look the same.
14
60160
2000
ื”ืŸ ื ืจืื•ืช ื“ื™ ืื•ืชื• ื”ื“ื‘ืจ
01:02
I've made him run into a wall here, over and over again.
15
62160
3000
ื’ืจืžืชื™ ืœื• ืœืจื•ืฅ ืœืชื•ืš ื”ืงื™ืจ, ืฉื•ื‘ ื•ืฉื•ื‘
01:05
And you can see he looks always the same.
16
65160
2000
ื•ืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ืฉื”ื•ื ื ืจืื” ื‘ื“ื™ื•ืง ืื•ืชื• ื”ื“ื‘ืจ
01:07
The reason for that is that these characters
17
67160
3000
ื”ืกื™ื‘ื” ืœื›ืš ื”ื™ื ืฉื”ื“ืžื•ื™ื•ืช ื”ืœืœื•
01:10
are actually not real characters.
18
70160
2000
ื”ืŸ ืœื ื‘ืืžืช ื“ืžื•ื™ื•ืช ืืžื™ืชื™ื•ืช
01:12
They are a graphical visualization of a character.
19
72160
4000
ื”ืŸ ื”ื“ืžื™ื” ื’ืจืคื™ืช ืฉืœ ื“ืžื•ืช
01:16
To produce these animations, an animator at a studio has to anticipate
20
76160
5000
ื›ื“ื™ ืœื”ื›ื™ืŸ ืื ื™ืžืฆื™ื•ืช ื›ืืœื•, ื”ืื ื™ืžื˜ื•ืจ ื‘ืกื˜ื•ื“ื™ื• ืฆืจื™ืš ืœื—ื–ื•ืช
01:21
what's going to happen in the actual game,
21
81160
3000
ืžื” ื™ืงืจื” ื‘ืžืฉื—ืง ื”ืืžื™ืชื™
01:24
and then has to animate that particular sequence.
22
84160
3000
ื•ืื– ืœื™ืฆื•ืจ ืืช ืจืฆืฃ ื”ืคืขื•ืœื•ืช ื”ืกืคืฆื™ืคื™ ื”ื–ื”
01:27
So, he or she sits down, animates it, and tries to anticipate what's going to happen,
23
87160
4000
ื›ืš ืฉื”ื•ื ืื• ื”ื™ื ื™ื•ืฉื‘ื™ื, ื™ื•ืฆืจื™ื ืืช ื”ืื ื™ืžืฆื™ื” ื”ื–ื•, ื•ืžื ืกื™ื ืœื—ื–ื•ืช ืžื” ื™ืงืจื” ื‘ืžืฉื—ืง.
01:31
and then these particular animations are just played back
24
91160
3000
ืœืื—ืจ ืžื›ืŸ ื”ืคืขื•ืœื•ืช ื”ืกืคืฆื™ืคื™ื•ืช ื”ืœืœื• ื—ื•ื–ืจื•ืช ืขืœ ืขืฆืžืŸ
01:34
at appropriate times in the computer game.
25
94160
3000
ื‘ื–ืžื ื™ื ื”ืชื•ืืžื™ื ืืช ื”ืคืขื•ืœื•ืช ื‘ืžื”ืœืš ื”ืžืฉื—ืง
01:37
Now, the result of that is that you can't have real interactivity.
26
97160
5000
ื”ืชื•ืฆืื” ื”ื™ื ืฉืื™ืŸ ื‘ืืžืช ืื™ื ื˜ืจืืงื˜ื™ื‘ื™ื•ืช ื‘ืžืฉื—ืง
01:42
All you have is animations that are played back
27
102160
3000
ื›ืš ืžืชืงื‘ืœ ืžืฆื‘ ืฉื‘ื• ื™ืฉื ืŸ ืคืขื•ืœื•ืช ืฉื—ื•ื–ืจื•ืช ืขืœ ืขืฆืžืŸ
01:45
at more or less the appropriate times.
28
105160
2000
ื‘ืžืจื•ื•ื—ื™ ื–ืžืŸ ืงื‘ื•ืขื™ื ืคื—ื•ืช ืื• ื™ื•ืชืจ
01:47
It also means that games aren't really going to be as surprising as they could be,
29
107160
5000
ื–ื” ืื•ืžืจ ืฉืžืฉื—ืงื™ ืžื—ืฉื‘ ืœื ื™ื”ื™ื• ืžืคืชื™ืขื™ื ื›ืžื• ืฉื”ื ื™ื›ื•ืœื™ื ืœื”ื™ื•ืช
01:52
because you only get out of it, at least in terms of the character,
30
112160
3000
ืžื›ื™ื•ื•ืŸ ืฉืืชื” ื™ื›ื•ืœ ืœืงื‘ืœ ืžื”ื, ืœืคื—ื•ืช ื‘ืžื•ื‘ืŸ ืฉืœ ื”ื“ืžื•ืช
01:55
what you actually put into it.
31
115160
2000
ืจืง ืžื” ืฉื”ื›ื ืกืช ืœื“ืžื•ืช
01:57
There's no real emergence there.
32
117160
2000
ืื™ืŸ ื”ืชื”ื•ื•ืช ืืžื™ืชื™ืช
01:59
And thirdly, as I said, most of the animations are very repetitive because of that.
33
119160
4000
ื•ืฉืœื™ืฉื™ืช, ื›ืžื• ืฉืืžืจืชื™, ืจื•ื‘ ื”ืื ื™ืžืฆื™ื•ืช ื—ื•ื–ืจื•ืช ืขืœ ืขืฆืžืŸ
02:03
Now, the only way to get around that
34
123160
2000
ื”ื“ืจืš ื”ื™ื—ื™ื“ื” ืฉื ื™ืชืŸ ืœืคืชื•ืจ ืืช ื–ื”
02:05
is to actually simulate the human body
35
125160
2000
ื–ื” ื‘ืขื–ืจืช ื”ื“ืžื™ื” ืืžืชื™ืช ืฉืœ ื”ื’ื•ืฃ ื”ืื ื•ืฉื™
02:07
and to simulate that bit of the nervous system of the brain that controls that body.
36
127160
5000
ื•ืœื“ืžื•ืช ืืช ื”ื—ืœืง ื‘ืžืขืจื›ืช ื”ืขืฆื‘ื™ื ื‘ืžื•ื— ืฉืฉื•ืœื˜ ื‘ื’ื•ืฃ
02:12
And maybe, if I could have you for a quick demonstration
37
132160
3000
ื•ืื•ืœื™, ืื ืื•ื›ืœ ืœื”ืฉืชืžืฉ ื‘ืš ืœื”ื“ื’ืžื” ื”ื‘ืื”
02:15
to show what the difference is --
38
135160
2000
ื›ื“ื™ ืœื”ืจืื•ืช ืืช ื”ื”ื‘ื“ืœ
02:17
because, I mean, it's very, very trivial.
39
137160
4000
ื›ื™ ื‘ืขืฆื ื–ื” ื ื•ืจื ื˜ืจื™ื•ื•ื™ืืœื™
02:21
If I push Chris a bit, like this, for example, he'll react to it.
40
141160
3000
ืื ืื ื™ ื“ื•ื—ืฃ ืืช ื›ืจื™ืก ื˜ื™ืคื”, ื›ื›ื”, ืœื“ื•ื’ืžื”, ื”ื•ื ื™ื’ื™ื‘ ืœื–ื”
02:24
If I push him from a different angle, he'll react to it differently,
41
144160
3000
ืื ืื“ื—ื•ืฃ ืื•ืชื• ืžื–ื•ื•ื™ืช ืฉื•ื ื”, ื”ื•ื ื™ื’ื™ื‘ ืื—ืจืช
02:27
and that's because he has a physical body,
42
147160
2000
ื•ื–ื” ื‘ื’ืœืœ ืฉื™ืฉ ืœื• ื’ื•ืฃ
02:29
and because he has the motor skills to control that body.
43
149160
3000
ื•ื‘ื’ืœืœ ืฉื™ืฉ ืœื• ื™ื›ื•ืœื•ืช ืžื•ื˜ื•ืจื™ื•ืช ืฉื™ืฉืœื˜ื• ื‘ื’ื•ืฃ ื”ื–ื”
02:32
It's a very trivial thing.
44
152160
2000
ื–ื” ืžืžืฉ ื˜ืจื™ื•ื•ื™ืืœื™
02:34
It's not something you get in computer games at the moment, at all.
45
154160
2000
ื–ื” ืœื ืžืฉื”ื• ืฉืงื™ื™ื ื›ื™ื•ื ื‘ืžืฉื—ืงื™ ืžื—ืฉื‘, ื‘ื›ืœืœ.
02:36
Thank you very much. Chris Anderson: That's it?
46
156160
2000
ืชื•ื“ื” ืจื‘ื”. ื›ืจื™ืก ืื ื“ืจืกื•ืŸ: ื–ื”ื•?
02:38
Torsten Reil: That's it, yes.
47
158160
2000
ื–ื”ื•, ื›ืŸ.
02:40
So, that's what we're trying to simulate --
48
160160
1000
ืื– ื–ื” ืžื” ืฉืื ื—ื ื• ืžื ืกื™ื ืœื“ืžื•ืช
02:41
not Chris specifically, I should say, but humans in general.
49
161160
4000
ืœื ืืช ื›ืจื™ืก ื‘ืื•ืคืŸ ืกืคืฆื™ืคื™ ืืœื ื‘ื ื™ ืื ื•ืฉ ื‘ื›ืœืœ
02:46
Now, we started working on this a while ago at Oxford University,
50
166160
5000
ื”ืชื—ืœื ื• ืœืขื‘ื•ื“ ืขืœ ื–ื” ื›ื‘ืจ ืœืคื ื™ ื–ืžืŸ ืžื” ื‘ืื•ื ื™ื‘ืจืกื™ื˜ืช ืื•ืงืกืคื•ืจื“
02:51
and we tried to start very simply.
51
171160
2000
ื•ื ื™ืกื™ื ื• ืœื”ืชื—ื™ืœ ื‘ืคืฉื˜ื•ืช
02:53
What we tried to do was teach a stick figure how to walk.
52
173160
3000
ืžื” ืฉื ื™ืกื™ื ื• ืœืขืฉื•ืช ื–ื” ืœืœืžื“ ื“ืžื•ืช ืžืžืงืœื•ืช ืœืœื›ืช
02:56
That stick figure is physically stimulated. You can see it here on the screen.
53
176160
3000
ื“ืžื•ืช ื”ืžืงืœื•ืช ื”ื–ื• ืžื“ื•ืžื” ืคื™ื–ื™ืช. ื ื™ืชืŸ ืœืจืื•ืช ืขืœ ื”ืžืกืš
02:59
So, it's subject to gravity, has joints, etc.
54
179160
3000
ื›ืš ืฉื”ื“ืžื•ืช ื ืชื•ื ื” ืœื›ื•ื— ื”ื›ื‘ื™ื“ื”, ื™ืฉ ืœื” ืžืคืจืงื™ื ื•ื›ื•ืœื™
03:02
If you just run the simulation, it will just collapse, like this.
55
182160
3000
ืื ื ืจื™ืฅ ืืช ื”ืกื™ืžื•ืœืฆื™ื” ื”ื“ืžื•ืช ืคืฉื•ื˜ ืชืชืžื•ื˜ื˜ ื›ื›ื”
03:05
The tricky bit is now to put an AI controller in it
56
185160
4000
ื”ื—ืœืง ื”ืžืกื•ื‘ืš ื”ื•ื ืœื”ื›ื ื™ืก ื‘ื“ืžื•ืช ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช ืฉืชืฉืœื•ื˜ ื‘ื”
03:09
that actually makes it work.
57
189160
2000
ืฉืชื•ื›ืœ ืœื’ืจื•ื ืœื” ืœืขื‘ื•ื“
03:11
And for that, we use the neural network, which we based on
58
191160
3000
ื•ืœืฆื•ืจืš ื›ืš ืื ื• ืžืฉืชืžืฉื™ื ื‘ืจืฉืช ืขืฆื‘ื™ืช ืฉืžื‘ื•ืกืกืช ืขืœ
03:14
that part of the nervous system that we have in our spine
59
194160
2000
ื”ื—ืœืง ื‘ืžืขืจื›ืช ื”ืขืฆื‘ื™ื ืฉื™ืฉ ืœื ื• ื‘ืขืžื•ื“ ื”ืฉื“ืจื”
03:16
that controls walking in humans.
60
196160
2000
ืฉืฉื•ืœื˜ ืขืœ ื”ื”ืœื™ื›ื” ื‘ื‘ื ื™ ืื“ื
03:18
It's called the central pattern generator.
61
198160
2000
ื–ื” ื ืงืจื ืžื—ื•ืœืœ ืชื‘ื ื™ืช ืžืจื›ื–ื™
03:20
So, we simulated that as well, and then the really tricky bit
62
200160
3000
ื›ืš ืฉืื ื• ืžื“ืžื™ื ืืช ื–ื” ื’ื, ื•ื”ื—ืœืง ื”ืžืกื•ื‘ืš ื‘ืืžืช
03:23
is to teach that network how to walk.
63
203160
2000
ื”ื•ื ืœืœืžื“ ืืช ื”ืจืฉืช ื”ืขืฆื‘ื™ืช ืœืœื›ืช
03:25
For that we used artificial evolution -- genetic algorithms.
64
205160
4000
ืœืฆื•ืจืš ื›ืš ื”ืฉืชืžืฉื ื• ื‘ืื‘ื•ืœื•ืฆื™ื” ืžืœืื›ื•ืชื™ืช. ืืœื’ื•ืจื™ืชืžื™ื ื’ื ื˜ื™ื™ื
03:29
We heard about those already yesterday,
65
209160
2000
ืฉืžืขื ื• ืขืœ ื›ืš ืืชืžื•ืœ
03:31
and I suppose that most of you are familiar with that already.
66
211160
3000
ื•ืื ื™ ืžื ื™ื— ืฉืจื•ื‘ื›ื ืžื›ื™ืจื™ื ืืช ื”ืžื•ืฉื’
03:34
But, just briefly, the concept is that
67
214160
2000
ืืš ื‘ืงืฆืจื”, ื”ืจืขื™ื•ืŸ ื”ื•ื
03:36
you create a large number of different individuals --
68
216160
3000
ืœื™ืฆื•ืจ ื›ืžื•ืช ื’ื“ื•ืœื” ืฉืœ ืคืจื˜ื™ื ืฉื•ื ื™ื,
03:39
neural networks, in this case --
69
219160
2000
ืจืฉืชื•ืช ืขืฆื‘ื™ื•ืช ื‘ืžืงืจื” ื”ื–ื”,
03:41
all of which are random at the beginning.
70
221160
2000
ืฉื›ื•ืœืŸ ืžืงืจื™ื•ืช ื‘ื”ืชื—ืœื”
03:43
You hook these up -- in this case, to the virtual muscles
71
223160
2000
ืืชื ืžื—ื‘ืจื™ื ืื•ืชืŸ ื™ื—ื“ ื‘ืžืงืจื” ื”ื–ื” ืœืฉืจื™ืจื™ื ืžืœืื›ื•ืชื™ื™ื
03:45
of that two-legged creature here --
72
225160
3000
ืฉืœ ื”ื™ืฆื•ืจ ื‘ืขืœ ืฉืชื™ ื”ืจื’ืœื™ื™ื ื”ื–ื”
03:48
and hope that it does something interesting.
73
228160
3000
ื•ืžืงื•ื•ื™ื ืฉื–ื” ื™ืขืฉื” ืžืฉื”ื• ืžืขื ื™ื™ืŸ
03:51
At the beginning, they're all going to be very boring.
74
231160
2000
ื‘ื”ืชื—ืœื” ื›ื•ืœื ืžืฉืขืžืžื™ื ืžืื“
03:53
Most of them won't move at all,
75
233160
2000
ืจื•ื‘ื ืœื ื™ื–ื•ื–ื• ื›ืœืœ
03:55
but some of them might make a tiny step.
76
235160
2000
ืื‘ืœ ื—ืœืงื ืื•ืœื™ ื™ืขืฉื• ืฆืขื“ ืงื˜ืŸ
03:57
Those are then selected by the algorithm,
77
237160
2000
ืืœื• ื ื‘ื—ืจื™ื ืขืœ ื™ื“ื™ ื”ืืœื’ื•ืจื™ืชื
03:59
reproduced with mutation and recombinations to introduce sex as well.
78
239160
4000
ืžื™ื•ืฆืจื•ืช ืžื—ื“ืฉ ืขืœ ื™ื“ื™ ืžื•ื˜ืฆื™ื” ื•ืฉื—ืœื•ืฃ ื›ื“ื™ ืœื”ื›ื ื™ืก ื’ื ืžื’ื“ืจ
04:03
And you repeat that process over and over again,
79
243160
2000
ื•ื—ื•ื–ืจื™ื ืขืœ ื”ืชื”ืœื™ืš ื”ื–ื” ืฉื•ื‘ ื•ืฉื•ื‘
04:05
until you have something that walks --
80
245160
2000
ืขื“ ืฉื™ืฉ ืœื›ื ืžืฉื”ื• ืฉื”ื•ืœืš
04:07
in this case, in a straight line, like this.
81
247160
2000
ื‘ืžืงืจื” ื”ื–ื”, ื‘ืงื• ื™ืฉืจ, ื›ื›ื”
04:09
So that was the idea behind this.
82
249160
2000
ืื– ื–ื” ื”ื™ื” ื”ืจืขื™ื•ืŸ ืžืื—ื•ืจื™ ื–ื”
04:11
When we started this, I set up the simulation one evening.
83
251160
3000
ื›ืฉื”ืชื—ืœื ื•, ืกื™ื“ืจืชื™ ืืช ื”ืชื•ื›ื ื” ืขืจื‘ ืื—ื“
04:14
It took about three to four hours to run the simulation.
84
254160
3000
ื–ื” ืœืงื— 3-4 ืฉืขื•ืช ืœื”ืจื™ืฅ ืืช ื”ืกื™ืžื•ืœืฆื™ื”.
04:17
I got up the next morning, went to the computer and looked at the results,
85
257160
4000
ื”ืชืขื•ืจืจืชื™ ื‘ื‘ื•ืงืจ ืฉืœืžื—ืจืช. ื”ืœื›ืชื™ ืœืžื—ืฉื‘ ื•ื”ื‘ื˜ืชื™ ื‘ืชื•ืฆืื•ืช
04:21
and was hoping for something that walked in a straight line,
86
261160
3000
ื•ืงื™ื•ื•ื™ืชื™ ืœื’ืœื•ืช ืžืฉื”ื• ืฉื”ื•ืœืš ื‘ืงื• ื™ืฉืจ
04:24
like I've just demonstrated,
87
264160
2000
ื›ืžื• ืฉื”ื“ื’ืžืชื™
04:26
and this is what I got instead.
88
266160
2000
ื–ื” ืžื” ืฉืงื™ื‘ืœืชื™ ื‘ืžืงื•ื
04:28
(Laughter)
89
268160
10000
(ืฆื—ื•ืง)
04:38
So, it was back to the drawing board for us.
90
278160
3000
ืื– ื—ื–ืจื ื• ืœืฉื•ืœื—ืŸ ื”ืชื›ื ื•ืŸ
04:42
We did get it to work eventually,
91
282160
3000
ืœื‘ืกื•ืฃ ื’ืจืžื ื• ืœื“ืžื•ืช ืœืขื‘ื•ื“
04:45
after tweaking a bit here and there.
92
285160
2000
ืื—ืจื™ ืฉืฉื™ื™ืคื ื• ืคื” ื•ืฉื
04:47
And this is an example of a successful evolutionary run.
93
287160
3000
ื•ื–ื•ื”ื™ ื“ื•ื’ืžื” ืœืชื”ืœื™ืš ืื‘ื•ืœื•ืฆื™ื•ื ื™ ืžื•ืฆืœื—
04:50
So, what you'll see in a moment is a very simple biped
94
290160
3000
ืžื” ืฉืชืจืื• ื‘ืขื•ื“ ืจื’ืข ื–ื• ื“ืžื•ืช ื‘ืขืœืช ืฉืชื™ ืจื’ืœื™ื™ื
04:53
that's learning how to walk using artificial evolution.
95
293160
3000
ืฉืœื•ืžื“ืช ืื™ืš ืœืœื›ืช ื‘ืขื–ืจืช ืื‘ื•ืœื•ืฆื™ื” ืžืœืื›ื•ืชื™ืช
04:56
At the beginning, it can't walk at all,
96
296160
2000
ื‘ื”ืชื—ืœื” ื”ื™ื ืœื ืžืฆืœื™ื—ื” ืœืœื›ืช
04:58
but it will get better and better over time.
97
298160
2000
ืืš ื”ื™ื ืชืฉืชืคืจ ืขื ื”ื–ืžืŸ
05:02
So, this is the one that can't walk at all.
98
302160
3000
ืื– ื–ื• ื“ืžื•ืช ืฉืœื ื”ื•ืœื›ืช
05:05
(Laughter)
99
305160
6000
(ืฆื—ื•ืง)
05:11
Now, after five generations of applying evolutionary process,
100
311160
3000
ืื—ืจื™ ื—ืžืฉื” ื“ื•ืจื•ืช ืฉืœ ื™ื™ืฉื•ื ืชื”ืœื™ืš ืื‘ื•ืœื•ืฆื™ื•ื ื™
05:14
the genetic algorithm is getting a tiny bit better.
101
314160
3000
ื”ืืœื’ื•ืจื™ืชื ื”ื’ื ื˜ื™ ืžืฉืชืคืจ ื˜ื™ืคื”
05:17
(Laughter)
102
317160
8000
(ืฆื—ื•ืง)
05:25
Generation 10 and it'll take a few steps more --
103
325160
2000
ื“ื•ืจ 10 ื™ืขืฉื” ืžืกืคืจ ืฆืขื“ื™ื ื ื•ืกืคื™ื
05:31
still not quite there.
104
331160
2000
ืขื“ื™ื™ืŸ ืœื ืฉื
05:34
But now, after generation 20, it actually walks in a straight line without falling over.
105
334160
5000
ืื‘ืœ ืขื›ืฉื™ื•, ืื—ืจื™ 20 ื“ื•ืจื•ืช, ื”ื•ื ื”ื•ืœืš ื‘ืงื• ื™ืฉืจ ื‘ืœื™ ืœื™ืคื•ืœ
05:40
That was the real breakthrough for us.
106
340160
3000
ื–ื• ื”ื™ื™ืชื” ืคืจื™ืฆืช ื”ื“ืจืš ื‘ืฉื‘ื™ืœื ื•
05:43
It was, academically, quite a challenging project,
107
343160
3000
ื–ื” ื”ื™ื” ืคืจื•ื™ื™ืงื˜ ืžืืชื’ืจ ื‘ืจืžื” ื”ืืงื“ืžื™ืช
05:46
and once we had reached that stage, we were quite confident
108
346160
3000
ื‘ืจื’ืข ืฉื”ื’ืขื ื• ืœืฉืœื‘ ื”ื–ื”, ื”ื™ื™ื ื• ื“ื™ ื‘ื˜ื•ื—ื™ื
05:49
that we could try and do other things as well with this approach --
109
349160
3000
ืฉืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื ืกื•ืช ืœืขืฉื•ืช ื“ื‘ืจื™ื ื ื•ืกืคื™ื ื‘ื’ื™ืฉื” ื”ื–ื•
05:52
actually simulating the body
110
352160
2000
ืžืžืฉ ืœื“ืžื•ืช ื’ื•ืฃ
05:54
and simulating that part of the nervous system that controls it.
111
354160
3000
ื•ืœื“ืžื•ืช ืืช ื”ื—ืœืง ื‘ืžืขืจื›ืช ื”ืขืฆื‘ื™ื ืฉืฉื•ืœื˜ ื‘ื•
05:57
Now, at this stage, it also became clear that this could be very exciting
112
357160
3000
ื‘ืฉืœื‘ ื–ื” ื”ื‘ื ื• ืฉื–ื” ื™ื›ื•ืœ ืœื”ื™ื•ืช ืžืื“ ืžืจื’ืฉ
06:00
for things like computer games or online worlds.
113
360160
3000
ืœืชื—ื•ืžื™ื ื›ืžื• ืžืฉื—ืงื™ ืžื—ืฉื‘ ืื• ืขื•ืœืžื•ืช ืื™ื ื˜ืจื ื˜ื™ื™ื
06:03
What you see here is the character standing there,
114
363160
2000
ืžื” ืฉืืชื ืจื•ืื™ื ื–ื• ื“ืžื•ืช ืฉืขื•ืžื“ืช
06:05
and there's an obstacle that we put in its way.
115
365160
2000
ื•ืื ื—ื ื• ื”ื ื—ื ื• ืžื›ืฉื•ืœ ื‘ื“ืจื›ื”
06:07
And what you see is, it's going to fall over the obstacle.
116
367160
5000
ืžื” ืฉื ื™ืชืŸ ืœืจืื•ืช, ื”ื™ื ื”ื•ืœื›ืช ืœื™ืคื•ืœ ืžื”ืžื›ืฉื•ืœ
06:12
Now, the interesting bit is, if I move the obstacle a tiny bit to the right,
117
372160
3000
ื”ื—ืœืง ื”ืžืขื ื™ื™ืŸ ื”ื•ื ืฉืื ืื ื™ ืื–ื™ื– ืืช ื”ืžื›ืฉื•ืœ ื˜ื™ืคื” ืœื™ืžื™ืŸ
06:15
which is what I'm doing now, here,
118
375160
2000
ืฉื–ื” ืžื” ืฉืื ื™ ืขื•ืฉื” ืขื›ืฉื™ื•
06:17
it will fall over it in a completely different way.
119
377160
4000
ื”ื“ืžื•ืช ืชื™ืคื•ืœ ื‘ืฆื•ืจื” ืฉื•ื ื” ืœื—ืœื•ื˜ื™ืŸ
06:24
And again, if you move the obstacle a tiny bit, it'll again fall differently.
120
384160
5000
ื•ืฉื•ื‘, ืื ืื–ื™ื– ืืช ื”ืžื›ืฉื•ืœ ื˜ื™ืคื”, ื”ื“ืžื•ืช ืชื™ืคื•ืœ ื‘ืื•ืคืŸ ืฉื•ื ื” ืœื—ืœื•ื˜ื™ืŸ
06:29
(Laughter)
121
389160
2000
(ืฆื—ื•ืง)
06:31
Now, what you see, by the way, at the top there,
122
391160
2000
ืžื” ืฉืืชื ืจื•ืื™ื ื“ืจืš ืื’ื‘ ื‘ื—ืœืง ื”ืขืœื™ื•ืŸ
06:33
are some of the neural activations being fed into the virtual muscles.
123
393160
3000
ื”ืŸ ื—ืœืง ืžื”ืคืขื™ืœื•ื™ื•ืช ื”ืขืฆื‘ื™ื•ืช ืฉืžื•ื–ื ื•ืช ืœืฉืจื™ืจื™ื ื”ื•ื™ืจื˜ื•ืืœื™ื
06:36
Okay. That's the video. Thanks.
124
396160
2000
ืื•.ืงื™. ื–ื” ื”ื•ื™ื“ืื•. ืชื•ื“ื”
06:38
Now, this might look kind of trivial, but it's actually very important
125
398160
3000
ื–ื” ื ืจืื” ื˜ืจื™ื•ื•ื™ืืœื™ ืืš ื–ื” ืžืื“ ื—ืฉื•ื‘
06:41
because this is not something you get at the moment
126
401160
2000
ื›ื™ ื–ื” ืœื ืžืฉื”ื• ืฉืงื™ื™ื ื”ื™ื•ื
06:43
in any interactive or any virtual worlds.
127
403160
2000
ื‘ืฉื•ื ืขื•ืœื ืžื“ื•ืžื” ืื• ืื™ื ื˜ืจืืงื˜ื™ื‘ื™
06:48
Now, at this stage, we decided to start a company and move this further,
128
408160
3000
ื‘ืฉืœื‘ ื–ื” ื”ื—ืœื˜ื ื• ืœืคืชื•ื— ื—ื‘ืจื” ื•ืœื”ืžืฉื™ืš ืœื”ืชืคืชื—
06:51
because obviously this was just a very simple, blocky biped.
129
411160
3000
ืžื›ื™ื•ื•ืŸ ืฉื–ื• ื”ื™ื™ืชื” ื“ืžื•ืช ืคืฉื•ื˜ื” ื•ืงื•ื‘ื™ื™ืชื™ืช
06:54
What we really wanted was a full human body.
130
414160
2000
ืžื” ืฉื‘ืืžืช ืจืฆื™ื ื• ื–ื” ื’ื•ืฃ ืื ื•ืฉื™ ืฉืœื
06:56
So we started the company.
131
416160
1000
ืื– ืคืชื—ื ื• ืืช ื”ื—ื‘ืจื”
06:57
We hired a team of physicists, software engineers and biologists
132
417160
5000
ืฉื›ืจื ื• ืฆื•ื•ืช ืฉืœ ืคื™ื–ื™ืงืื™ื, ืžื”ื ื“ืกื™ ืชื•ื›ื ื” ื•ื‘ื™ื•ืœื•ื’ื™ื
07:02
to work on this, and the first thing we had to work on
133
422160
3000
ืฉื™ืขื‘ื“ื• ืขืœ ื–ื”, ื•ื”ื“ื‘ืจ ื”ืจืืฉื•ืŸ ืฉืขื‘ื“ื ื• ืขืœื™ื•
07:05
was to create the human body, basically.
134
425160
4000
ื”ื™ื” ืœื™ืฆื•ืจ ื’ื•ืฃ ืื ื•ืฉื™
07:09
It's got to be relatively fast, so you can run it on a normal machine,
135
429160
3000
ื–ื” ื”ื™ื” ื—ื™ื™ื‘ ืœื”ื™ืขืฉื•ืช ืžื”ืจ ื›ื“ื™ ืฉื ื•ื›ืœ ืœื‘ื“ื•ืง ืื•ืชื• ื‘ืžื›ื•ื ื” ืจื’ื™ืœื”
07:12
but it's got to be accurate enough, so it looks good enough, basically.
136
432160
3000
ืืš ื–ื” ื”ื™ื” ืฆืจื™ืš ืœื”ื™ื•ืช ืžื“ื•ื™ืง ืžืกืคื™ืง ื›ื“ื™ ืฉื–ื” ื™ื™ืจืื” ืžืกืคื™ืง ื˜ื•ื‘
07:15
So we put quite a bit of biomechanical knowledge into this thing,
137
435160
3000
ืื– ื”ื›ื ืกื ื• ื“ื™ ื”ืจื‘ื” ืžื™ื“ืข ื‘ื™ื• ืžื›ืื ื™ ื‘ื“ื‘ืจ ื”ื–ื”
07:18
and tried to make it as realistic as possible.
138
438160
4000
ื•ื ื™ืกื™ื ื• ืœื™ืฆื•ืจ ื“ืžื•ืช ืืžื™ืชื™ืช ื›ื›ืœ ืฉื ื™ืชืŸ
07:22
What you see here on the screen right now
139
442160
2000
ืžื” ืฉืืชื ืจื•ืื™ื ืขืœ ื”ืžืกืš ื›ืจื’ืข
07:24
is a very simple visualization of that body.
140
444160
2000
ื–ื• ื”ื“ืžื™ื” ืคืฉื•ื˜ื” ืฉืœ ื”ื’ื•ืฃ ื”ืื ื•ืฉื™
07:26
I should add that it's very simple to add things like hair, clothes, etc.,
141
446160
4000
ืืฆื™ื™ืŸ ืฉื–ื” ื“ื™ ืคืฉื•ื˜ ืœื”ื•ืกื™ืฃ ื“ื‘ืจื™ื ื›ืžื• ืฉื™ืขืจ, ื‘ื’ื“ื™ื ื•ื›ื•ืœื™.
07:30
but what we've done here is use a very simple visualization,
142
450160
3000
ืืš ืื ื—ื ื• ื‘ื—ืจื ื• ืœื”ืฉืชืžืฉ ื‘ื”ื“ืžื™ื” ืคืฉื•ื˜ื”
07:33
so you can concentrate on the movement.
143
453160
2000
ื›ืš ืฉื ื•ื›ืœ ืœื”ืชืจื›ื– ืขืœ ื”ืชื ื•ืขื”
07:35
Now, what I'm going to do right now, in a moment,
144
455160
3000
ืขื›ืฉื™ื• ืžื” ืฉืื ื™ ื”ื•ืœืš ืœืขืฉื•ืช ื‘ืขื•ื“ ืจื’ืข
07:38
is just push this character a tiny bit and we'll see what happens.
145
458160
3000
ื–ื” ืœื“ื—ื•ืฃ ืืช ื”ื“ืžื•ืช ื˜ื™ืคื”, ื•ื ืจืื” ืžื” ื™ืงืจื”
07:46
Nothing really interesting, basically.
146
466160
2000
ืœื ืžืฉื”ื• ืžืขื ื™ื™ืŸ
07:48
It falls over, but it falls over like a rag doll, basically.
147
468160
3000
ื”ื“ืžื•ืช ื ื•ืคืœืช ื›ืžื• ื‘ื•ื‘ื”.
07:51
The reason for that is that there's no intelligence in it.
148
471160
3000
ื”ืกื™ื‘ื” ื”ื™ื ืฉืื™ืŸ ืœื“ืžื•ืช ืื™ื ื˜ืœื™ื’ื ืฆื™ื”
07:54
It becomes interesting when you put artificial intelligence into it.
149
474160
4000
ื–ื” ื ืขืฉื” ืžืขื ื™ื™ืŸ ื›ืฉืžื›ื ื™ืกื™ื ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช ืœื“ืžื•ืช
07:58
So, this character now has motor skills in the upper body --
150
478160
4000
ืœื“ืžื•ืช ื”ื–ื• ื™ืฉ ื™ื›ื•ืœื•ืช ืžื•ื˜ื•ืจื™ื•ืช ื‘ื—ืœืง ื”ื’ื•ืฃ ื”ืขืœื™ื•ืŸ
08:02
nothing in the legs yet, in this particular one.
151
482160
2000
ืœื“ืžื•ืช ื”ื–ื• ืื™ืŸ ื™ื›ื•ืœื•ืช ื‘ืจื’ืœื™ื™ื
08:04
But what it will do -- I'm going to push it again.
152
484160
3000
ืืš ืžื” ืฉื”ื“ืžื•ืช ืชืขืฉื”, ืื ื™ ืื“ื—ื•ืฃ ืื•ืชื” ืฉื•ื‘
08:07
It will realize autonomously that it's being pushed.
153
487160
2000
ื”ื“ืžื•ืช ืชื‘ื™ืŸ ื‘ืื•ืคืŸ ืขืฆืžืื™ ืฉื”ื™ื ื ื“ื—ืคืช
08:09
It's going to stick out its hands.
154
489160
2000
ื”ื™ื ืชืฉืœื— ืืช ื™ื“ื™ื”
08:11
It's going to turn around into the fall, and try and catch the fall.
155
491160
3000
ื”ื™ื ืชืกืชื•ื‘ื‘ ืœื›ื™ื•ื•ืŸ ื”ื ืคื™ืœื” ื•ืชื ืกื” ืœื‘ืœื•ื ืืช ื”ื ืคื™ืœื”
08:20
So that's what you see here.
156
500160
2000
ืื– ื–ื” ืžื” ืฉืืชื ืจื•ืื™ื
08:22
Now, it gets really interesting
157
502160
2000
ืขื›ืฉื™ื• ื–ื” ื ืขืฉื” ืžืžืฉ ืžืขื ื™ื™ืŸ
08:24
if you then add the AI for the lower part of the body as well.
158
504160
4000
ืื ื ื›ื ื™ืก ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช ืœื—ืœืง ื”ื’ื•ืฃ ื”ืชื—ืชื•ืŸ
08:28
So here, we've got the same character.
159
508160
2000
ื›ืš ืฉื™ืฉ ืœื ื• ืืช ืื•ืชื” ื”ื“ืžื•ืช
08:30
I'm going to push it a bit harder now,
160
510160
2000
ืื ื™ ืื“ื—ื•ืฃ ืื•ืชื” ืงืฆืช ื™ื•ืชืจ ื—ื–ืง
08:32
harder than I just pushed Chris.
161
512160
2000
ื™ื•ืชืจ ื—ื–ืง ืžืžื” ืฉื“ื—ืคืชื™ ืืช ื›ืจื™ืก
08:34
But what you'll see is -- it's going to receive a push now from the left.
162
514160
4000
ืžื” ืฉืชืจืื• ื–ื” ืฉื”ื“ืžื•ืช ืชืงื‘ืœ ื“ื—ื™ืคื” ืžืฉืžืืœ
08:41
What you see is it takes steps backwards,
163
521160
2000
ืชื™ืงื— ืžืกืคืจ ืฆืขื“ื™ื ืื—ื•ืจื”
08:43
it tries to counter-balance,
164
523160
2000
ืชื ืกื” ืœืื–ืŸ ืืช ืขืฆืžื”
08:45
it tries to look at the place where it thinks it's going to land.
165
525160
4000
ื•ืชื ืกื” ืœื”ื‘ื™ื˜ ืœืžืงื•ื ืฉื‘ื• ื”ื™ื ื—ื•ืฉื‘ืช ืฉื”ื™ื ืชื™ืคื•ืœ
08:49
I'll show you this again.
166
529160
2000
ืืจืื” ื–ืืช ืฉื•ื‘
08:51
And then, finally hits the floor.
167
531160
3000
ื•ืœื‘ืกื•ืฃ ื”ื™ื ืคื•ื’ืขืช ื‘ืจืฆืคื”
08:55
Now, this becomes really exciting
168
535160
3000
ืขื›ืฉื™ื• ื–ื” ื ืขืฉื” ืžืžืฉ ืžืœื”ื™ื‘
08:58
when you push that character in different directions, again, just as I've done.
169
538160
5000
ืื ืชื“ื—ืคื• ืืช ื”ื“ืžื•ืช ืžื›ื™ื•ื•ื ื™ื ืฉื•ื ื™ื ื›ืžื• ืฉืขืฉื™ืชื™
09:03
That's something that you cannot do right now.
170
543160
4000
ื–ื” ืžืฉื”ื• ืฉืœื ื ื™ืชืŸ ืœืขืฉื•ืช ื”ื™ื•ื
09:07
At the moment, you only have empty computer graphics in games.
171
547160
3000
ื›ื™ื•ื ื™ืฉ ืจืง ื’ืจืคื™ืงื•ืช ืจื™ืงื•ืช ื‘ืžืฉื—ืงื™ื
09:10
What this is now is a real simulation. That's what I want to show you now.
172
550160
3000
ืขื›ืฉื™ื• ื–ื• ืกื™ืžื•ืœืฆื™ื” ืืžื™ืชื™ืช, ื–ื” ืžื” ืฉืื ื™ ืจื•ืฆื” ืœื”ืจืื•ืช ืœื›ื ืขื›ืฉื™ื•
09:13
So, here's the same character with the same behavior I've just shown you,
173
553160
3000
ื–ื• ืื•ืชื” ื”ื“ืžื•ืช ืขื ืื•ืชื” ื”ื”ืชื ื”ื’ื•ืช ื›ืžื• ืฉื”ืจืื™ืชื™ ืœื›ื
09:16
but now I'm just going to push it from different directions.
174
556160
2000
ืืš ืขื›ืฉื™ื• ืื ื™ ืื“ื—ื•ืฃ ืื•ืชื” ืžื›ื™ื•ื•ืŸ ืฉื•ื ื”
09:18
First, starting with a push from the right.
175
558160
2000
ืจืืฉื™ืช, ืืชื—ื™ืœ ืขื ื“ื—ื™ืคื” ืžื™ืžื™ืŸ
09:23
This is all slow motion, by the way, so we can see what's going on.
176
563160
3000
ื–ื” ื‘ื”ื™ืœื•ืš ืื™ื˜ื™ ื›ืš ืฉื ื•ื›ืœ ืœืจืื•ืช ืžื” ืงื•ืจื”
09:26
Now, the angle will have changed a tiny bit,
177
566160
3000
ืขื›ืฉื™ื• ื”ื–ื•ื•ื™ืช ื”ืฉืชื ืชื” ื˜ื™ืคื”
09:29
so you can see that the reaction is different.
178
569160
4000
ื›ืš ืฉืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ืฉื”ืชื’ื•ื‘ื” ืฉื•ื ื”
09:33
Again, a push, now this time from the front.
179
573160
3000
ืฉื•ื‘, ื“ื—ื™ืคื” ืžืœืคื ื™ื ื”ืคืขื
09:37
And you see it falls differently.
180
577160
2000
ื•ืืชื ืจื•ืื™ื ืฉื”ื™ื ื ื•ืคืœืช ืื—ืจืช
09:39
And now from the left --
181
579160
2000
ื•ืขื›ืฉื™ื• ืžืฉืžืืœ
09:43
and it falls differently.
182
583160
2000
ื•ื”ื™ื ื ื•ืคืœืช ื‘ืฆื•ืจื” ืฉื•ื ื”
09:45
That was really exciting for us to see that.
183
585160
2000
ื–ื” ื”ื™ื” ืžืžืฉ ืžืœื”ื™ื‘ ื‘ืฉื‘ื™ืœื ื• ืœืจืื•ืช ืืช ื–ื”
09:47
That was the first time we've seen that.
184
587160
2000
ื–ื• ื”ื™ื™ืชื” ื”ืคืขื ื”ืจืืฉื•ื ื” ืฉืจืื™ื ื• ืืช ื–ื”
09:49
This is the first time the public sees this as well,
185
589160
2000
ื–ื• ื”ืคืขื ื”ืจืืฉื•ื ื” ืฉื”ืฆื™ื‘ื•ืจ ืจื•ืื” ื–ืืช
09:51
because we have been in stealth mode.
186
591160
2000
ื›ื™ ื”ื™ื™ื ื• ื—ืฉืื™ื™ื ื‘ื ื•ื’ืข ืœื–ื”
09:53
I haven't shown this to anybody yet.
187
593160
2000
ืœื ื”ืจืื™ืชื™ ืืช ื–ื” ืœืืฃ ืื—ื“ ืขื“ื™ื™ืŸ
09:55
Now, just a fun thing:
188
595160
2000
ืขื›ืฉื™ื•, ืžืฉื”ื• ื ื—ืžื“
09:57
what happens if you put that character --
189
597160
2000
ืžื” ืงื•ืจื” ืื ืชืฉื™ื ืืช ื”ื“ืžื•ืช
09:59
this is now a wooden version of it, but it's got the same AI in it --
190
599160
2000
ื–ื• ืจืง ื“ืžื•ืช ืขืฅ ืืš ื™ืฉ ื‘ื” ืืช ืื•ืชื” ื”ืื ื˜ื™ืœื™ื’ื ืฆื™ื” ื”ืžืœืื›ื•ืชื™ืช
10:01
but if you put that character on a slippery surface, like ice.
191
601160
2000
ืืš ืื ืชื ื™ื— ืืช ื”ื“ืžื•ืช ืขืœ ืžืฉื˜ื— ื—ืœืงืœืง ื›ืžื• ืงืจื—
10:03
We just did that for a laugh, just to see what happens.
192
603160
3000
ืขืฉื™ื ื• ืืช ื–ื” ืจืง ื‘ืฉื‘ื™ืœ ื”ืฆื—ื•ืง, ืจืง ื›ื“ื™ ืœืจืื•ืช ืžื” ืงื•ืจื”
10:06
(Laughter)
193
606160
1000
(ืฆื—ื•ืง)
10:07
And this is what happens.
194
607160
2000
ื•ื–ื” ืžื” ืฉืงื•ืจื”
10:09
(Laughter)
195
609160
3000
(ืฆื—ื•ืง)
10:12
(Applause)
196
612160
3000
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
10:15
It's nothing we had to do about this.
197
615160
2000
ืœื ื”ื™ื™ื ื• ืฆืจื™ื›ื™ื ืœืขืฉื•ืช ื“ื‘ืจ
10:17
We just took this character that I just talked about,
198
617160
2000
ืคืฉื•ื˜ ืœืงื—ื ื• ืืช ื”ื“ืžื•ืช ืฉื”ืจื’ืข ื“ื™ื‘ืจื ื• ืขืœื™ื”
10:19
put it on a slippery surface, and this is what you get out of it.
199
619160
3000
ืœื”ื ื™ื— ืื•ืชื” ืขืœ ืžืฉื˜ื— ื—ืœืง ื•ื–ื” ืžื” ืฉืžืงื‘ืœื™ื
10:22
And that's a really fascinating thing about this approach.
200
622160
3000
ื•ื–ื” ืžื” ืฉืžืขื ื™ื™ืŸ ื‘ื’ื™ืฉื” ื”ื–ื•
10:26
Now, when we went to film studios and games developers
201
626160
3000
ืขื›ืฉื™ื• ืฉื”ืœื›ื ื• ืœืžืคืชื—ื™ ืžืฉื—ืงื™ ืžื—ืฉื‘ ื•ื—ื‘ืจื•ืช ืกืจื˜ื™ื
10:29
and showed them that technology, we got a very good response.
202
629160
3000
ื•ื”ืจืื™ื ื• ืœื”ื ืืช ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ื”ื–ื•, ืงื™ื‘ืœื ื• ืชื’ื•ื‘ื” ื˜ื•ื‘ื” ืžืื“
10:32
And what they said was, the first thing they need immediately is virtual stuntmen.
203
632160
4000
ื•ืžื” ืฉื”ื ืืžืจื• ื”ื™ื”, ืฉื”ื“ื‘ืจ ื”ืจืืฉื•ืŸ ืฉื”ื ื–ืงื•ืงื™ื ืœื• ื–ื” ืคืขืœื•ืœืŸ ื•ื™ืจื˜ื•ืืœื™
10:36
Because stunts are obviously very dangerous, they're very expensive,
204
636160
4000
ืžื›ื™ื•ื•ืŸ ืฉืคืขืœื•ืœื™ื ื”ื ื‘ื‘ื™ืจื•ืจ ื“ื‘ืจ ืžืกื•ื›ืŸ ื•ื”ื ืžืื“ ื™ืงืจื™ื
10:40
and there are a lot of stunt scenes that you cannot do, obviously,
205
640160
2000
ื•ื™ืฉื ืŸ ื”ืžื•ืŸ ืกืฆื ื•ืช ืคืขืœื•ืœื™ื ืฉืœื ื ื™ืชืŸ ืœืขืฉื•ืช
10:42
because you can't really allow the stuntman to be seriously hurt.
206
642160
3000
ืžื›ื™ื•ื•ืŸ ืฉืืชื” ืœื ื™ื›ื•ืœ ืœืคื’ื•ืข ื‘ืคืขืœื•ืœืŸ
10:45
So, they wanted to have a digital version of a stuntman
207
645160
3000
ืื– ื”ื ืจืฆื• ืคืขืœื•ืœืŸ ื“ื™ื’ื™ื˜ืœื™
10:48
and that's what we've been working on for the past few months.
208
648160
2000
ื•ืขืœ ื–ื” ืื ื—ื ื• ืขื•ื‘ื“ื™ื ื‘ื—ื•ื“ืฉื™ื ื”ืื—ืจื•ื ื™ื
10:50
And that's our first product that we're going to release in a couple of weeks.
209
650160
5000
ื•ื–ื” ื”ืžื•ืฆืจ ื”ืจืืฉื•ืŸ ืฉืื ื—ื ื• ืžืชื›ื•ื•ื ื™ื ืœื”ื•ืฆื™ื ื‘ืขื•ื“ ืžืกืคืจ ื—ื•ื“ืฉื™ื
10:55
So, here are just a few very simple scenes of the guy just being kicked.
210
655160
5000
ืื– ืœืคื ื™ื›ื ืžืกืคืจ ืกืฆื ื•ืช ืฉืœ ื‘ื—ื•ืจ ืฉืžืงื‘ืœ ื‘ืขื™ื˜ื”
11:00
That's what people want. That's what we're giving them.
211
660160
2000
ื–ื” ืžื” ืฉืื ืฉื™ื ืจื•ืฆื™ื, ื–ื” ืžื” ืฉืกื™ืคืงื ื• ืœื”ื
11:02
(Laughter)
212
662160
7000
(ืฆื—ื•ืง)
11:09
You can see, it's always reacting.
213
669160
2000
ืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช ืฉื”ื“ืžื•ืช ืชืžื™ื“ ืžื’ื™ื‘ื”
11:11
This is not a dead body. This is a body who basically, in this particular case,
214
671160
4000
ื–ื• ืœื ื’ื•ืคื”, ื–ื”ื• ื’ื•ืฃ ืฉื‘ื—ืœืง ื”ืกืคืฆื™ืคื™ ื”ื–ื”
11:15
feels the force and tries to protect its head.
215
675160
2000
ืžืจื’ื™ืฉ ืืช ื”ื›ื•ื— ื•ืžื ืกื” ืœื”ื’ืŸ ืขืœ ืจืืฉื•
11:17
Only, I think it's quite a big blow again.
216
677160
2000
ืื ื™ ื—ื•ืฉื‘ ืฉื–ื• ืžื›ื” ื“ื™ ืจืฆื™ื ื™ืช ืฉื•ื‘
11:19
You feel kind of sorry for that thing,
217
679160
2000
ืืชื” ื“ื™ ืžืจื—ื ืขืœ ื”ื“ืžื•ืช
11:21
and we've seen it so many times now that
218
681160
2000
ื•ืจืื™ื ื• ืืช ื–ื” ื›ืœ ื›ืš ื”ืจื‘ื” ืคืขืžื™ื
11:23
we don't really care any more.
219
683160
2000
ืฉืœื ืžืžืฉ ืื›ืคืช ืœื ื• ื™ื•ืชืจ
11:25
(Laughter)
220
685160
1000
(ืฆื—ื•ืง)
11:26
There are much worse videos than this, by the way, which I have taken out, but ...
221
686160
4000
ื™ืฉ ืงื˜ืขื™ ื•ื™ื“ืื• ื”ืจื‘ื” ื™ื•ืชืจ ื’ืจื•ืขื™ื ืžื–ื” ื“ืจืš ืื’ื‘ ืฉื”ื•ืฆืืชื™
11:31
Now, here's another one.
222
691160
2000
ื›ืืŸ ืขื•ื“ ืื—ื“
11:33
What people wanted as a behavior was to have an explosion,
223
693160
4000
ืžื” ืฉืื ืฉื™ื ืจืฆื• ื–ื• ืชื’ื•ื‘ื” ืœืคื™ืฆื•ืฅ
11:37
a strong force applied to the character,
224
697160
2000
ื›ื•ื— ื—ื–ืง ืฉืคื•ื’ืข ื‘ื“ืžื•ืช
11:39
and have the character react to it in midair.
225
699160
2000
ื•ืฉื”ื“ืžื•ืช ืชื’ื™ื‘ ืœืคื™ืฆื•ืฅ ื‘ืขื•ื“ื” ื‘ืื•ื™ืจ
11:41
So that you don't have a character that looks limp,
226
701160
2000
ื›ืš ืฉื”ื“ืžื•ืช ืœื ืชื™ืจืื” ืžื•ื’ื‘ืœืช
11:43
but actually a character that you can use in an action film straight away,
227
703160
3000
ืืœื ื“ืžื•ืช ืฉืชื•ื›ืœ ืœื”ืฉืชืžืฉ ื‘ื” ื‘ืกืจื˜ ืืงืฉืŸ
11:46
that looks kind of alive in midair as well.
228
706160
2000
ื“ืžื•ืช ืฉื ืจืื™ืช ื—ื™ื” ื’ื ื‘ืขื•ื“ื” ืขืคื”
11:48
So this character is going to be hit by a force,
229
708160
2000
ืื– ื”ื“ืžื•ืช ื”ื–ืืช ื”ื•ืœื›ืช ืœื”ื™ืคื’ืข ืžื›ื•ื— ื’ื“ื•ืœ
11:50
it's going to realize it's in the air,
230
710160
2000
ื”ื™ื ื”ื•ืœื›ืช ืœื”ื‘ื™ืŸ ืฉื”ื™ื ื ืžืฆืืช ื‘ืื•ื™ืจ
11:52
and it's going to try and, well,
231
712160
3000
ื•ื”ื“ืžื•ืช ืชื ืกื”, ื•ื‘ื›ืŸ
11:55
stick out its arm in the direction where it's landing.
232
715160
2000
ืœืคืจื•ืก ืืช ื™ื“ื™ื” ืœื›ื™ื•ื•ืŸ ื”ื ื—ื™ืชื”
11:59
That's one angle; here's another angle.
233
719160
3000
ื–ื• ื–ื•ื•ื™ืช ืื—ืช, ื”ื ื” ืขื•ื“ ื–ื•ื•ื™ืช
12:02
We now think that the realism we're achieving with this
234
722160
2000
ืื ื—ื ื• ืžืืžื™ื ื™ื ืฉืจืžืช ื”ืจื™ืืœื™ื–ื ืฉืื ื—ื ื• ืžืฉื™ื’ื™ื
12:04
is good enough to be used in films.
235
724160
2000
ื˜ื•ื‘ื” ืžืกืคื™ืง ืœืฉื™ืžื•ืฉ ื‘ืชืขืฉื™ื™ืช ื”ืกืจื˜ื™ื
12:06
And let's just have a look at a slightly different visualization.
236
726160
3000
ื‘ื•ืื• ื ืจืื” ื”ื“ืžื™ื” ืงืฆืช ืฉื•ื ื”
12:09
This is something I just got last night
237
729160
2000
ื–ื” ืžืฉื”ื• ืฉืงื™ื‘ืœืชื™ ืืชืžื•ืœ ื‘ืœื™ืœื”
12:11
from an animation studio in London, who are using our software
238
731160
3000
ืžืกื˜ื•ื“ื™ื• ืื ื™ืžืฆื™ื” ื‘ืœื•ื ื“ื•ืŸ, ืฉืžืฉืชืžืฉ ื‘ืชื•ื›ื ื” ืฉืœื ื•
12:14
and experimenting with it right now.
239
734160
2000
ื•ืžื ืกื” ืื•ืชื” ื‘ื™ืžื™ื ืืœื•
12:16
So this is exactly the same behavior that you saw,
240
736160
3000
ืื– ื–ื• ืื•ืชื” ื”ืชื ื”ื’ื•ืช ืฉืจืื™ืชื
12:19
but in a slightly better rendered version.
241
739160
4000
ืจืง ื‘ื’ืจืกื” ืžื”ื•ืงืฆืขืช ื™ื•ืชืจ
12:23
So if you look at the character carefully,
242
743160
3000
ืื– ืื ืชื‘ื™ื˜ื• ื‘ื“ืžื•ืช ืœืขื•ืžืง
12:26
you see there are lots of body movements going on,
243
746160
2000
ืชืจืื• ืฉื™ืฉ ื”ืžื•ืŸ ืชื ื•ืขื•ืช ื’ื•ืฃ ืฉืžืชืจื—ืฉื•ืช
12:28
none of which you have to animate like in the old days.
244
748160
2000
ืœื ืžืฉื”ื• ืฉื“ื•ืžื” ืœืื ื™ืžืฆื™ื” ืฉืœ ื™ืžื™ื ืขื‘ืจื•
12:30
Animators had to actually animate them.
245
750160
2000
ื‘ื™ืžื™ื ืฉืื ื™ืžื˜ื•ืจื™ื ื”ื™ื• ืฆืจื™ื›ื™ื ืœื™ืฆื•ืจ ืืช ื”ืชื ื•ืขื•ืช ื”ืœืœื•
12:32
This is all happening automatically in the simulation.
246
752160
2000
ื›ืœ ื”ืชื ื•ืขื•ืช ืงื•ืจื•ืช ืื•ื˜ื•ืžื˜ื™ืช ื‘ืžื”ืœืš ื”ืกื™ืžื•ืœืฆื™ื”
12:34
This is a slightly different angle,
247
754160
2000
ื–ื• ื–ื•ื•ื™ืช ืฉื•ื ื”
12:39
and again a slow motion version of this.
248
759160
2000
ื•ืฉื•ื‘, ื’ืจืกื” ื‘ื”ื™ืœื•ืš ืื™ื˜ื™
12:41
This is incredibly quick. This is happening in real time.
249
761160
4000
ื–ื” ืžื”ื™ืจ ืœื”ืคืœื™ื. ื–ื” ืงื•ืจื” ื‘ื–ืžืŸ ืืžืช
12:45
You can run this simulation in real time, in front of your eyes,
250
765160
2000
ืืคืฉืจ ืœื”ืจื™ืฅ ืืช ื”ืกื™ืžื•ืœืฆื™ื” ื‘ื–ืžืŸ ืืžืช, ืœื ื’ื“ ืขื™ื ื™ืš
12:47
change it, if you want to, and you get the animation straight out of it.
251
767160
3000
ืœืฉื ื•ืช ืื•ืชื” ืื ืชืจืฆื”, ื•ืชืงื‘ืœ ืืช ื”ืื ื™ืžืฆื™ื” ื™ืฉื™ืจื•ืช
12:50
At the moment, doing something like this by hand
252
770160
2000
ื›ื™ื•ื, ืœืขืฉื•ืช ื“ื‘ืจ ื›ื–ื” ื‘ืื•ืคืŸ ื™ื“ื ื™
12:52
would take you probably a couple of days.
253
772160
2000
ื™ื™ืงื— ืœืš ืžืกืคืจ ื™ืžื™ื
12:55
This is another behavior they requested.
254
775160
3000
ื–ื• ื”ืชื ื”ื’ื•ืช ื ื•ืกืคืช ืฉื”ื ื‘ื™ืงืฉื•
12:58
I'm not quite sure why, but we've done it anyway.
255
778160
2000
ืื ื™ ืœื ืžืžืฉ ื‘ื˜ื•ื— ืœืžื” ืื‘ืœ ืขืฉื™ื ื• ืืช ื–ื” ื‘ื›ืœ ืžืงืจื”
13:00
It's a very simple behavior that shows you the power of this approach.
256
780160
2000
ื–ื• ื”ืชื ื”ื’ื•ืช ืคืฉื•ื˜ื” ืฉืžืจืื” ืืช ื”ื›ื•ื— ืฉืœ ื”ื’ื™ืฉื” ื”ื–ื•
13:02
In this case, the character's hands
257
782160
2000
ื‘ืžืงืจื” ื”ื–ื” ื™ื“ื™ื” ืฉืœ ื”ื“ืžื•ืช
13:04
are fixed to a particular point in space,
258
784160
2000
ืžืงื•ื‘ืขื•ืช ืœื ืงื•ื“ื” ืžืกื•ื™ืžืช ื‘ื—ืœืœ
13:06
and all we've told the character to do is to struggle.
259
786160
3000
ื•ื›ืœ ืžื” ืฉื‘ื™ืงืฉื ื• ืžื”ื“ืžื•ืช ืœืขืฉื•ืช ื–ื” ืœื”ื™ืื‘ืง
13:09
And it looks organic. It looks realistic.
260
789160
3000
ื•ื–ื” ื ืจืื” ืื•ืจื’ื ื™, ื ืจืื” ืžืฆื™ืื•ืชื™
13:12
You feel kind of sorry for the guy.
261
792160
2000
ืืคืฉืจ ืœืจื—ื ืขืœ ื”ื‘ื—ื•ืจ
13:14
It's even worse -- and that is another video I just got last night --
262
794160
3000
ื–ื” ืืคื™ืœื• ื’ืจื•ืข ื™ื•ืชืจ, ื•ื–ื” ืขื•ื“ ื•ื™ื“ืื• ืฉืงื™ื‘ืœืชื™ ืืชืžื•ืœ ื‘ืœื™ืœื”
13:17
if you render that a bit more realistically.
263
797160
2000
ืื ืชื—ื“ื“ ืืช ื–ื” ื•ืชื”ืคื•ืš ืืช ื–ื” ืœืžืฆื™ืื•ืชื™ ื™ื•ืชืจ
13:23
Now, I'm showing this to you just to show you
264
803160
2000
ืขื›ืฉื™ื• ืื ื™ ืืจืื” ืœื›ื ืืช ื–ื”
13:25
how organic it actually can feel, how realistic it can look.
265
805160
2000
ืื™ืš ื–ื” ื™ื›ื•ืœ ืœื”ืจื’ื™ืฉ ืื•ืจื’ื ื™, ื›ืžื” ื–ื” ืžืจื’ื™ืฉ ืžืฆื™ืื•ืชื™
13:27
And this is all a physical simulation of the body,
266
807160
3000
ื•ื–ื” ืกื™ืžื•ืœืฆื™ื” ืคื™ื–ื™ืช ืฉืœ ื’ื•ืฃ ื”ืื“ื
13:30
using AI to drive virtual muscles in that body.
267
810160
3000
ื‘ืขื–ืจืช ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช ืฉืžื ื™ืขื” ืฉืจื™ืจื™ื ื‘ื’ื•ืฃ
13:35
Now, one thing which we did for a laugh was
268
815160
3000
ืขื•ื“ ื“ื‘ืจ ืฉืขืฉื™ื ื• ืจืง ื‘ืฉื‘ื™ืœ ื”ืฆื—ื•ืง
13:38
to create a slightly more complex stunt scene,
269
818160
2000
ื›ื“ื™ ืœื™ืฆื•ืจ ืกืฆื ืช ืคืขืœื•ืœื™ื ืžื•ืจื›ื‘ืช
13:40
and one of the most famous stunts is the one where James Bond
270
820160
3000
ื•ืื—ื“ ื”ืคืขืœื•ืœื™ื ื”ืžืคื•ืจืกืžื™ื ื‘ื™ื•ืชืจ ื”ื•ื ื‘ื—ืœืง ื‘ื• ื’'ื™ื™ืžืก ื‘ื•ื ื“
13:43
jumps off a dam in Switzerland and then is caught by a bungee.
271
823160
4000
ืงื•ืคืฅ ืžืกื›ืจ ื‘ืฉื•ื•ื™ืฅ ื•ืื– ื ืชืคืก ื‘ื—ื‘ืœ ื‘ืื ื’'ื™
13:48
Got a very short clip here.
272
828160
2000
ืงืœื™ืค ืงืฆืจ
13:54
Yes, you can just about see it here.
273
834160
2000
ื›ืŸ, ื ื™ืชืŸ ืœืจืื•ืช ืืช ื–ื” ื›ืืŸ
13:56
In this case, they were using a real stunt man. It was a very dangerous stunt.
274
836160
3000
ื‘ืžืงืจื” ื”ื–ื” ื”ื ื”ืฉืชืžืฉื• ื‘ืคืขืœื•ืœืŸ ืืžื™ืชื™. ืคืขืœื•ืœ ืžืื“ ืžืกื•ื›ืŸ
13:59
It was just voted, I think in the Sunday Times, as one of the most impressive stunts.
275
839160
3000
ื‘ืกืื ื“ื™ื™ ื˜ื™ื™ืžืก ื‘ื—ืจื• ื‘ื• ื›ืื—ื“ ื”ืคืขืœื•ืœื™ื ื”ืžืจืฉื™ืžื™ื ื‘ื™ื•ืชืจ
14:02
Now, we've just tried and -- looked at our character and asked ourselves,
276
842160
3000
ื”ื‘ื˜ื ื• ื‘ื“ืžื•ืช ืฉืœื ื• ื•ืฉืืœื ื• ืืช ืขืฆืžื ื•
14:05
"Can we do that ourselves as well?"
277
845160
2000
ื”ืื ื’ื ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืขืฉื•ืช ื–ืืช?
14:07
Can we use the physical simulation of the character,
278
847160
2000
ื”ืื ื ื•ื›ืœ ืœื”ืฉืชืžืฉ ื‘ืกื™ืžื•ืœืฆื™ื” ื”ืคื™ื–ื™ืช ืฉืœ ื”ื“ืžื•ืช
14:09
use artificial intelligence,
279
849160
2000
ืœื”ืฉืชืžืฉ ื‘ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช
14:11
put that artificial intelligence into the character,
280
851160
2000
ืœื”ื›ื ื™ืก ืืช ื”ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ื”ืžืœืื›ื•ืชื™ืช ื”ื–ื• ืœื“ืžื•ืช ืฉืœื ื•
14:13
drive virtual muscles, simulate the way he jumps off the dam,
281
853160
4000
ืœื”ื ื™ืข ืฉืจื™ืจื™ื ื•ื™ืจื˜ื•ืืœื™ื, ืœื“ืžื•ืช ืืช ื”ื“ืจืš ื‘ื” ื”ื•ื ืงื•ืคืฅ ืžื”ืกื›ืจ
14:17
and then skydive afterwards,
282
857160
2000
ื•ืื– ืœืฆื ื•ื— ืฆื ื™ื—ื” ื—ื•ืคืฉื™ืช
14:19
and have him caught by a bungee afterwards?
283
859160
2000
ื•ืœื’ืจื•ื ืœื—ื‘ืœ ื”ื‘ืื ื’'ื™ ืœืชืคื•ืก ืื•ืชื•?
14:21
We did that. It took about altogether just two hours,
284
861160
3000
ืขืฉื™ื ื• ืืช ื–ื”, ื–ื” ืœืงื— ืœื ื• ื‘ืขืจืš ืฉืขืชื™ื™ื
14:24
pretty much, to create the simulation.
285
864160
2000
ื›ื“ื™ ืœื™ืฆื•ืจ ืืช ื”ืกื™ืžื•ืœืฆื™ื”
14:26
And that's what it looks like, here.
286
866160
2000
ื•ื›ืš ื–ื” ื ืจืื”
14:37
Now, this could do with a bit more work. It's still very early stages,
287
877160
3000
ืืคืฉืจ ืขื•ื“ ืœืขื‘ื•ื“ ืขืœ ื–ื”. ืืœื• ืขื“ื™ื™ืŸ ืฉืœื‘ื™ื ืจืืฉื•ื ื™ื
14:40
and we pretty much just did this for a laugh,
288
880160
2000
ื•ื“ื™ ืขืฉื™ื ื• ืืช ื–ื” ื‘ืฉื‘ื™ืœ ื”ืฆื—ื•ืง
14:42
just to see what we'd get out of it.
289
882160
2000
ืจืง ืœืจืื•ืช ืžื” ื™ืฆื ืœื ื• ืžื–ื”
14:44
But what we found over the past few months
290
884160
2000
ืื‘ืœ ืžื” ืฉื’ื™ืœื™ื ื• ื‘ื—ื•ื“ืฉื™ื ื”ืื—ืจื•ื ื™ื
14:46
is that this approach -- that we're pretty much standard upon --
291
886160
3000
ืฉื”ื’ื™ืฉื” ื”ื–ื• ืฉืื ื—ื ื• ื“ื™ ืกื’ื•ืจื™ื ืขืœื™ื”
14:49
is incredibly powerful.
292
889160
2000
ื—ื–ืงื” ืžืื“
14:51
We are ourselves surprised what you actually get out of the simulations.
293
891160
4000
ืื ื—ื ื• ื“ื™ ืžื•ืคืชืขื™ื ืžืžื” ืฉืืคืฉืจ ืœืงื‘ืœ ืžื”ืกื™ืžื•ืœืฆื™ื•ืช
14:55
There's very often very surprising behavior that you didn't predict before.
294
895160
4000
ื”ืจื‘ื” ืคืขืžื™ื ืื ื—ื ื• ื ืชืงืœื™ื ื‘ื”ืชื ื”ื’ื•ื™ื•ืช ืžืคืชื™ืขื•ืช ืฉืœื ืฆืคื™ื ื• ืžืจืืฉ
14:59
There's so many things we can do with this right now.
295
899160
2000
ื™ืฉ ื›ืœ ื›ืš ื”ืจื‘ื” ื“ื‘ืจื™ื ืฉื ื™ืชืŸ ืœืขืฉื•ืช ื‘ืขื–ืจืช ื”ืชื•ื›ื ื”
15:01
The first thing, as I said, is going to be virtual stuntmen.
296
901160
3000
ื”ื“ื‘ืจ ื”ืจืืฉื•ืŸ ื›ืžื• ืฉืืžืจืชื™ ื–ื” ืœืขืฉื•ืช ืคืขืœื•ืœืŸ ื•ื™ืจื˜ื•ืืœื™
15:04
Several studios are using this software now to produce virtual stuntmen,
297
904160
4000
ืžืกืคืจ ืื•ืœืคื ื™ื ืžืฉืชืžืฉื™ื ื‘ืชื•ื›ื ื” ื”ื–ื• ื›ื“ื™ ืœื™ื™ืฆืจ ืคืขืœื•ืœืŸ ื•ื™ืจื˜ื•ืืœื™
15:08
and they're going to hit the screen quite soon, actually,
298
908160
2000
ื•ื”ื ืขื•ืœื™ื ืœืืงืจื ื™ื ื‘ืงืจื•ื‘ ืžืื“
15:10
for some major productions.
299
910160
2000
ื‘ื›ืžื” ื”ืคืงื•ืช ื’ื“ื•ืœื•ืช
15:12
The second thing is video games.
300
912160
3000
ื”ื“ื‘ืจ ื”ืฉื ื™ ื”ื•ื ืžืฉื—ืงื™ ืžื—ืฉื‘
15:15
With this technology, video games will look different and they will feel very different.
301
915160
4000
ื‘ืขื–ืจืช ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ื”ื–ื•, ืžืฉื—ืงื™ ืžื—ืฉื‘ ื™ืจืื• ืฉื•ื ื” ื•ื™ืจื’ื™ืฉื• ืฉื•ื ื”
15:19
For the first time, you'll have actors that really feel very interactive,
302
919160
3000
ื‘ืคืขื ื”ืจืืฉื•ื ื” ื™ื”ื™ื• ื“ืžื•ื™ื•ืช ืฉื™ืจื’ื™ืฉื• ืื™ื ื˜ืจืืงื˜ื™ื‘ื™ื•ืช
15:22
that have real bodies that really react.
303
922160
2000
ื•ื‘ืขืœื•ืช ื’ื•ืฃ ืฉืžื’ื™ื‘ ื‘ืืžืช
15:24
I think that's going to be incredibly exciting.
304
924160
3000
ืื ื™ ื—ื•ืฉื‘ ืฉื–ื” ื™ื”ื™ื” ืžืœื”ื™ื‘ ืžืื“
15:27
Probably starting with sports games,
305
927160
2000
ื‘ื˜ื— ื ืชื—ื™ืœ ื‘ืžืฉื—ืงื™ ืกืคื•ืจื˜
15:29
which are going to become much more interactive.
306
929160
2000
ืฉื™ื”ืคื›ื• ืœืื™ื ื˜ืจืืงื˜ื™ื‘ื™ื™ื ื”ืจื‘ื” ื™ื•ืชืจ
15:31
But I particularly am really excited
307
931160
1000
ืื‘ืœ ืื ื™ ื ืœื”ื‘ ื‘ืžื™ื•ื—ื“
15:32
about using this technology in online worlds,
308
932160
3000
ืžื”ืฉื™ืžื•ืฉ ื‘ืชื•ื›ื ื” ื‘ืขื•ืœืžื•ืช ืื•ื ืœื™ื™ืŸ
15:35
like there, for example, that Tom Melcher has shown us.
309
935160
3000
ื›ืžื• ืžื” ืฉื˜ื•ื ืžืœืฆ'ืจ ื”ืจืื” ืœื ื•
15:38
The degree of interactivity you're going to get
310
938160
2000
ืžื™ื“ืช ื”ืื™ื ื˜ืจืืงื˜ื™ื‘ื™ื•ืช ืฉื ืงื‘ืœ
15:40
is totally different, I think, from what you're getting right now.
311
940160
3000
ืชื”ื™ื” ืฉื•ื ื” ืœื—ืœื•ื˜ื™ืŸ, ืื ื™ ื—ื•ืฉื‘, ืžืžื” ืฉืื ื—ื ื• ืžืงื‘ืœื™ื ืขื›ืฉื™ื•
15:44
A third thing we are looking at and very interested in is simulation.
312
944160
4000
ื”ื“ื‘ืจ ื”ืฉืœื™ืฉื™ ืฉืื ื—ื ื• ืžืชืขื ื™ื™ื ื™ื ื‘ื• ื”ื•ื ื”ืกื™ืžื•ืœืฆื™ื”
15:49
We've been approached by several simulation companies,
313
949160
2000
ืžืกืคืจ ื—ื‘ืจื•ืช ืกื™ืžื•ืœืฆื™ื” ื›ื‘ืจ ื ื™ื’ืฉื• ืืœื™ื ื•
15:51
but one project we're particularly excited about, which we're starting next month,
314
951160
3000
ืื‘ืœ ืคืจื•ื™ืงื˜ ืื—ื“ ืžืœื”ื™ื‘ ื‘ืžื™ื•ื—ื“, ืฉื ืชื—ื™ืœ ื‘ื—ื•ื“ืฉ ื”ื‘ื
15:54
is to use our technology -- and in particular, the walking technology --
315
954160
4000
ืฉื™ืžื•ืฉ ื‘ื˜ื›ื ื•ืœื•ื’ื™ื” ืฉืœื ื• ื•ื‘ืžื™ื•ื—ื“ ืืช ื˜ื›ื ื•ืœื•ื’ื™ื™ืช ื”ื”ืœื™ื›ื”
15:58
to help aid surgeons who work on children with cerebral palsy,
316
958160
4000
ืœืขื–ื•ืจ ืœืžื ืชื—ื™ื ืฉืขื•ื‘ื“ื™ื ืขื ื™ืœื“ื™ื ืขื ืฉื™ืชื•ืง ืžื•ื—ื™ืŸ
16:02
to predict the outcome of operations on these children.
317
962160
3000
ืœื—ื–ื•ืช ืืช ืชื•ืฆืื•ืช ื”ื ื™ืชื•ื—ื™ื ื•ื”ืฉืคืขืชื ืขืœ ื”ื™ืœื“ื™ื
16:05
As you probably know,
318
965160
2000
ื›ืžื• ืฉืืชื ื•ื“ืื™ ื™ื•ื“ืขื™ื
16:07
it's very difficult to predict what the outcome of an operation is
319
967160
3000
ืงืฉื” ืžืื“ ืœื—ื–ื•ืช ืชื•ืฆืื” ืฉืœ ื ื™ืชื•ื—
16:10
if you try and correct the gait.
320
970160
2000
ืื ืืชื” ืžื ืกื” ืœืชืงืŸ ื”ืœื™ื›ื”
16:12
The classic quote is, I think, it's unpredictable at best,
321
972160
3000
ื”ืฆื™ื˜ื•ื˜ ื”ืงืœืืกื™ ื”ื•ื "ื–ื” ืœื ืฆืคื•ื™ ื‘ืžืงืจื” ื”ื˜ื•ื‘"
16:15
is what people think right now, is the outcome.
322
975160
3000
ื–ื” ืžื” ืฉืื ืฉื™ื ื—ื•ืฉื‘ื™ื ื›ื™ื•ื ืฉืชื”ื™ื” ื”ืชื•ืฆืื”
16:18
Now, what we want to do with our software is allow our surgeons to have a tool.
323
978160
4000
ืžื” ืฉืื ื—ื ื• ืจื•ืฆื™ื ืœืขืฉื•ืช ืขื ื”ืชื•ื›ื ื” ืฉืœื ื• ื–ื” ืœืชืช ืœืžื ืชื—ื™ื ื›ืœื™
16:22
We're going to simulate the gait of a particular child
324
982160
3000
ืื ื—ื ื• ื”ื•ืœื›ื™ื ืœื“ืžื•ืช ื”ืœื™ื›ื” ืฉืœ ื™ืœื“ ืžืกื•ื™ื
16:25
and the surgeon can then work on that simulation
325
985160
3000
ื•ืื– ื”ืžื ืชื— ื™ื›ื•ืœ ืœืขื‘ื•ื“ ืขืœ ื”ืกื™ืžื•ืœืฆื™ื” ื”ื–ื•
16:28
and try out different ways to improve that gait,
326
988160
2000
ื•ืœื ืกื•ืช ื“ืจื›ื™ื ืฉื•ื ื•ืช ืœืฉืคืจ ืืช ื”ื”ืœื™ื›ื” ื”ื–ื•
16:30
before he actually commits to an actual surgery.
327
990160
3000
ืœืคื ื™ ืฉื”ื•ื ืžื‘ืฆืข ื ื™ืชื•ื— ื‘ืคื•ืขืœ
16:33
That's one project we're particularly excited about,
328
993160
2000
ื–ื” ืคืจื•ื™ืงื˜ ืื—ื“ ืฉืžืื“ ืžืœื”ื™ื‘ ืื•ืชื ื•
16:35
and that's going to start next month.
329
995160
2000
ื•ื–ื” ื™ื—ืœ ื‘ืขื•ื“ ื›ื—ื•ื“ืฉ
16:39
Just finally, this is only just the beginning.
330
999160
3000
ืœื‘ืกื•ืฃ, ื–ื• ืจืง ื”ื”ืชื—ืœื”
16:42
We can only do several behaviors right now.
331
1002160
2000
ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื“ืžื•ืช ืจืง ืžืกืคืจ ื”ืชื ื”ื’ื•ื™ื•ืช ืขื›ืฉื™ื•
16:44
The AI isn't good enough to simulate a full human body.
332
1004160
3000
ื”ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ื”ืžืœืื›ื•ืชื™ืช ืœื ื˜ื•ื‘ื” ืžืกืคื™ืง ืœื“ืžื•ืช ื’ื•ืฃ ืื ื•ืฉื™ ืฉืœื
16:47
The body yes, but not all the motor skills that we have.
333
1007160
3000
ืืช ื”ื’ื•ืฃ ื›ืŸ ืืš ืœื ืืช ื›ืœ ื”ื™ื›ื•ืœื•ืช ื”ืžื•ื˜ื•ืจื™ื•ืช ืฉื™ืฉ ืœื ื•
16:50
And, I think, we're only there if we can have something like ballet dancing.
334
1010160
3000
ื•ืื ื™ ื—ื•ืฉื‘ ืฉืื ื—ื ื• ื ื”ื™ื” ืฉื ืื ื ื•ื›ืœ ืœื“ืžื•ืช ืžืฉื”ื• ื›ืžื• ืจืงื“ื ื™ืช ื‘ืœื˜
16:53
Right now, we don't have that
335
1013160
2000
ื›ืจื’ืข ืื™ืŸ ืœื ื• ืืช ื–ื”
16:55
but I'm very sure that we will be able to do that at some stage.
336
1015160
2000
ืื‘ืœ ืื ื™ ื“ื™ ื‘ื˜ื•ื— ืฉื ื•ื›ืœ ืœืขืฉื•ืช ืืช ื–ื” ื‘ื‘ื•ื ื”ื™ื•ื
16:57
We do have one unintentional dancer actually,
337
1017160
3000
ื”ืืžืช ื”ื™ื, ืฉื™ืฉ ืœื ื• ื‘ืœื™ ื›ื•ื•ื ื” ืจืงื“ืŸ
17:00
the last thing I'm going to show you.
338
1020160
2000
ื”ื“ื‘ืจ ื”ืื—ืจื•ืŸ ืฉืื ื™ ืืจืื” ืœื›ื
17:02
This was an AI contour that was produced and evolved --
339
1022160
3000
ื–ื• ืื™ื ื˜ืœื™ื’ื ืฆื™ื” ืžืœืื›ื•ืชื™ืช ืฉื”ืชืคืชื—ื”
17:05
half-evolved, I should say -- to produce balance, basically.
340
1025160
3000
ื›ื“ืื™ ืฉืื’ื™ื“, ื—ืฆื™ ื”ืชืคืชื—ื” ื›ื“ื™ ืœื™ื™ืฆืจ ืฉื™ื•ื•ื™ ืžืฉืงืœ
17:08
So, you kick the guy and the guy's supposed to counter-balance.
341
1028160
3000
ืื– ื‘ืขื˜ื ื• ื‘ื‘ื—ื•ืจ ื•ื”ื‘ื—ื•ืจ ื”ื™ื” ืืžื•ืจ ืœื”ืชืื–ืŸ
17:11
That's what we thought was going to come out of this.
342
1031160
3000
ื–ื” ืžื” ืฉื—ืฉื‘ื ื• ืฉื™ืฆื ืœื ื•
17:14
But this is what emerged out of it, in the end.
343
1034160
2000
ืืš ื–ื” ืžื” ืฉื™ืฆื ืœื ื• ื‘ืกื•ืคื• ืฉืœ ื“ื‘ืจ
17:17
(Music)
344
1037160
10000
(ืžื•ื–ื™ืงื”)
17:27
Bizarrely, this thing doesn't have a head. I'm not quite sure why.
345
1047160
3000
ื‘ืื•ืคืŸ ืžื•ื–ืจ, ืื™ืŸ ืœื“ื‘ืจ ื”ื–ื” ืจืืฉ. ืื ื™ ืœื ื‘ื˜ื•ื— ืœืžื”
17:31
So, this was not something we actually put in there.
346
1051160
2000
ื–ื” ืœื ืžืฉื”ื• ืฉื”ื›ื ืกื ื• ืœืฉื ื‘ืืžืช
17:33
He just started to create that dance himself.
347
1053160
4000
ื”ื•ื ืคืฉื•ื˜ ื”ืชื—ื™ืœ ืœื™ืฆื•ืจ ืืช ื”ืจื™ืงื•ื“ ื”ื–ื” ื‘ืขืฆืžื•
17:37
He's actually a better dancer than I am, I have to say.
348
1057160
3000
ื”ืืžืช ืฉื”ื•ื ืจืงื“ืŸ ื™ื•ืชืจ ื˜ื•ื‘ ืžืžื ื™, ืื ื™ ื—ื™ื™ื‘ ืœืฆื™ื™ืŸ
17:41
And what you see after a while --
349
1061160
2000
ื•ืžื” ืฉืืชื ืจื•ืื™ื ืœืื—ืจ ื–ืžืŸ ืžื”
17:43
I think he even goes into a climax right at the end.
350
1063160
2000
ืื ื™ ื—ื•ืฉื‘ ืฉื”ื•ื ืžื’ื™ืข ืœืฉื™ื ืžืžืฉ ื‘ืกื•ืฃ
17:49
And I think -- there you go.
351
1069160
3000
ื•ืื ื™ ื—ื•ืฉื‘, ื”ื ื” ืœื›ื.
17:52
(Laughter)
352
1072160
2000
(ืฆื—ื•ืง)
17:54
So, that all happened automatically. We didn't put that in there.
353
1074160
2000
ื›ืœ ื–ื” ืงืจื” ืื•ื˜ื•ืžื˜ื™ืช. ืื ื—ื ื• ืœื ื ื’ืขื ื•
17:56
That's just the simulation creating this itself, basically.
354
1076160
3000
ื–ื• ื”ืกื™ืžื•ืœืฆื™ื” ืฉื™ื•ืฆืจืช ืืช ื–ื” ื‘ืขืฆืžื”
17:59
So it's just --
355
1079160
2000
ื›ืš ืฉื”ื™ื
18:01
(Applause)
356
1081160
1000
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
18:02
Thanks.
357
1082160
2000
ืชื•ื“ื”
18:05
Not quite John Travolta yet, but we're working on that as well,
358
1085160
3000
ืœื ืžืžืฉ ื’'ื•ืŸ ื˜ืจื‘ื•ืœื˜ื”, ืื‘ืœ ืื ื—ื ื• ืขื•ื‘ื“ื™ื ื’ื ืขืœ ื–ื”
18:08
so thanks very much for your time.
359
1088160
2000
ืื– ืชื•ื“ื” ืจื‘ื” ืขืœ ื–ืžื ื›ื
18:10
Thanks.
360
1090160
1000
ืชื•ื“ื”
18:11
(Applause)
361
1091160
1000
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
18:12
CA: Incredible. That was really incredible.
362
1092160
2000
ื›ืจื™ืก: ืžื“ื”ื™ื. ื–ื” ื”ื™ื” ืžืžืฉ ืžื“ื”ื™ื
18:14
TR: Thanks.
363
1094160
1000
ืชื•ื“ื”
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7