TED2003
Torsten Reil: Animate characters by evolving them
애니메이션 제작을 위한 Torsten Reil의 생물학 연구
Filmed:
Readability: 3.5
363,842 views
Torsten Reil은 생물학이 어떻게 애니메이션속 사람들을 자연스럽게 보이도록 할 수 있는지 가상 인간의 뼈, 근육, 신경들을 구축함으로써 이야기합니다. 2003년 TED에서 강연한 내용을 통해 그가 GTA4라는 곳에서 한 일들을 살펴봅니다
Torsten Reil - Animating neurobiologist
By coding computer simulations with biologically modeled nervous systems, Torsten Reil and his company NaturalMotion breathe life into the animated characters inhabiting the most eye-poppingly realistic games and movies around. Full bio
By coding computer simulations with biologically modeled nervous systems, Torsten Reil and his company NaturalMotion breathe life into the animated characters inhabiting the most eye-poppingly realistic games and movies around. Full bio
Double-click the English transcript below to play the video.
00:15
I'm going to talk about a technology that we're developing at Oxford now,
0
3000
4000
저는 지금 옥스포드에서 개발중인 기술에 관해 얘기할까 합니다.
00:19
that we think is going to change the way that
1
7000
3000
이 기술은 앞으로 컴퓨터 게임과 할리우드 영화가
00:22
computer games and Hollywood movies are being made.
2
10000
3000
만들어 지는 방법을 바꿀꺼라고 생각합니다.
00:26
That technology is simulating humans.
3
14000
3000
이 기술은 바로 인간의 행동을 흉내내는 것입니다.
00:29
It's simulated humans with a simulated body
4
17000
3000
인간의 몸과 그 몸을 제어하는 신경을 가진
00:32
and a simulated nervous system to control that body.
5
20000
3000
애니메이션속 인간에 관한 것입니다.
00:36
Now, before I talk more about that technology,
6
24000
3000
이 기술에 대해 얘기하기 전에,
00:39
let's have a quick look at what human characters look like
7
27000
3000
컴퓨터 게임에 나오는 사람은 어떻게 보이는지
00:42
at the moment in computer games.
8
30000
3000
잠시 보시죠.
00:45
This is a clip from a game called "Grand Theft Auto 3."
9
33000
3000
이것은 Grand Theft Auto3라는 게임에 나오는 한 장면입니다.
00:48
We already saw that briefly yesterday.
10
36000
2000
저는 어제 잠깐 봤구요.
00:50
And what you can see is -- it is actually a very good game.
11
38000
3000
지금 보시는 것은 사실 상당히 좋은 게임입니다.
00:53
It's one of the most successful games of all time.
12
41000
3000
최고의 게임중 하나죠.
00:56
But what you'll see is that all the animations in this game are very repetitive.
13
44000
4000
그런데 보시면 아시겠지만 게임속의 모든 애니메이션이 매우 반복적입니다.
01:00
They pretty much look the same.
14
48000
2000
모두들 비슷하게 보이죠.
01:02
I've made him run into a wall here, over and over again.
15
50000
3000
이사람을 벽에 부딪치게 해 보았습니다. 반복해서.
01:05
And you can see he looks always the same.
16
53000
2000
언제나 같은 움직임을 보실 수 있죠.
01:07
The reason for that is that these characters
17
55000
3000
그 이유는 바로 여기 게임속 캐릭터들이
01:10
are actually not real characters.
18
58000
2000
실제 사람이 아니기 때문입니다.
01:12
They are a graphical visualization of a character.
19
60000
4000
사람을 컴퓨터 그래픽으로 시각화한 것 뿐입니다.
01:16
To produce these animations, an animator at a studio has to anticipate
20
64000
5000
이런 영상을 스튜디오에서 만들기 위해서는
01:21
what's going to happen in the actual game,
21
69000
3000
제작자는 실제 게임에서 어떤 일들이 벌어질지 예측하고,
01:24
and then has to animate that particular sequence.
22
72000
3000
연속동작들을 그려내야 하죠.
01:27
So, he or she sits down, animates it, and tries to anticipate what's going to happen,
23
75000
4000
그래서 제작자는 자리에 앉아 어떤일이 생길지 고민하여 그려보고
01:31
and then these particular animations are just played back
24
79000
3000
그렇게 만들어진 영상이 게임속에서 특정시점에
01:34
at appropriate times in the computer game.
25
82000
3000
반복적으로 재현되는 것 뿐이죠.
01:37
Now, the result of that is that you can't have real interactivity.
26
85000
5000
현실감 있는 상호작용이 없게 됩니다.
01:42
All you have is animations that are played back
27
90000
3000
똑같은 모습들이 계속적으로 반복되는
01:45
at more or less the appropriate times.
28
93000
2000
모습들만 보게되는 거죠
01:47
It also means that games aren't really going to be as surprising as they could be,
29
95000
5000
이런점들때문에 게임들이 별로 놀라울게 없게 됩니다.
01:52
because you only get out of it, at least in terms of the character,
30
100000
3000
왜냐하면 애니메이션속 사람은
01:55
what you actually put into it.
31
103000
2000
입력한 대로만 움직이기 때문입니다.
01:57
There's no real emergence there.
32
105000
2000
급박한게 없다고 할까요.
01:59
And thirdly, as I said, most of the animations are very repetitive because of that.
33
107000
4000
그래서 결국 반복적인 행동만 하게 되죠.
02:03
Now, the only way to get around that
34
111000
2000
이걸 해결하는 유일한 방법은
02:05
is to actually simulate the human body
35
113000
2000
바로 사람의 몸과
02:07
and to simulate that bit of the nervous system of the brain that controls that body.
36
115000
5000
그 몸을 제어하는 뇌신경을 흉내내는 수밖에 없습니다.
02:12
And maybe, if I could have you for a quick demonstration
37
120000
3000
차이가 뭔지 여러분께 잠깐 보여드리자면,
02:15
to show what the difference is --
38
123000
2000
그 차이는
02:17
because, I mean, it's very, very trivial.
39
125000
4000
아주 사소한 겁니다.
02:21
If I push Chris a bit, like this, for example, he'll react to it.
40
129000
3000
제가 Chirs를 조금 밀면 반응하죠.
02:24
If I push him from a different angle, he'll react to it differently,
41
132000
3000
제가 다른 방향에서 밀면 다르게 반응하죠
02:27
and that's because he has a physical body,
42
135000
2000
그건 Chris가 실제 몸을 가지고
02:29
and because he has the motor skills to control that body.
43
137000
3000
그 몸을 통제하는 근육도 가지고 있기 때문입니다.
02:32
It's a very trivial thing.
44
140000
2000
아주 사소하지만
02:34
It's not something you get in computer games at the moment, at all.
45
142000
2000
게임속에서는 전혀 찾아볼 수가 없습니다.
02:36
Thank you very much. Chris Anderson: That's it?
46
144000
2000
고마와요. 크리스앤더슨: 이게 다인가요?
02:38
Torsten Reil: That's it, yes.
47
146000
2000
Torsten: 네 그게 다입니다.
02:40
So, that's what we're trying to simulate --
48
148000
1000
우리가 흉내내고자 하는것은
02:41
not Chris specifically, I should say, but humans in general.
49
149000
4000
Chris가 아니라 일반적인 사람이라고 할 수 있죠.
02:46
Now, we started working on this a while ago at Oxford University,
50
154000
5000
옥스포드 대학에서 얼마전부터 이 연구를
02:51
and we tried to start very simply.
51
159000
2000
아주 간단한것 부터 시작했습니다.
02:53
What we tried to do was teach a stick figure how to walk.
52
161000
3000
막대기 캐릭터에게 걷는 방법을 가르쳤보았습니다.
02:56
That stick figure is physically stimulated. You can see it here on the screen.
53
164000
3000
여기 스크린속 막대기는 실제 사람을 흉내내고 있는데 보이시죠.
02:59
So, it's subject to gravity, has joints, etc.
54
167000
3000
중력을 받으며, 관절들도 있죠.
03:02
If you just run the simulation, it will just collapse, like this.
55
170000
3000
시뮬레이션을 돌리면 막대기는 이렇게 주저앉습니다.
03:05
The tricky bit is now to put an AI controller in it
56
173000
4000
이제 좀 복잡한데 인공지능 제어기능을 추가하면
03:09
that actually makes it work.
57
177000
2000
실제 걸을 수 있습니다.
03:11
And for that, we use the neural network, which we based on
58
179000
3000
이를 위해서 신경회로를 활용했는데,
03:14
that part of the nervous system that we have in our spine
59
182000
2000
우리 척추에 있는 신경의 일부로
03:16
that controls walking in humans.
60
184000
2000
걷는 동작을 통제하죠.
03:18
It's called the central pattern generator.
61
186000
2000
소위 중추유형발생기(CPG)라고 합니다.
03:20
So, we simulated that as well, and then the really tricky bit
62
188000
3000
그것도 흉내내도록 했는데, 어려운 점은
03:23
is to teach that network how to walk.
63
191000
2000
신경회로가 걷도록 가르치는 겁니다.
03:25
For that we used artificial evolution -- genetic algorithms.
64
193000
4000
이를 위해서 유전알고리즘을 이용하여 인공적으로 진화하도록 했죠
03:29
We heard about those already yesterday,
65
197000
2000
어제 이 내용은 이미 들었고,
03:31
and I suppose that most of you are familiar with that already.
66
199000
3000
아마 대부분 익숙하실 듯 합니다.
03:34
But, just briefly, the concept is that
67
202000
2000
간략히 개념을 설명하자만
03:36
you create a large number of different individuals --
68
204000
3000
많은 인공의 사람을 만들고
03:39
neural networks, in this case --
69
207000
2000
여기서 신경회로는
03:41
all of which are random at the beginning.
70
209000
2000
임의로 시작합니다.
03:43
You hook these up -- in this case, to the virtual muscles
71
211000
2000
이제 가상의 두발달린 모양의
03:45
of that two-legged creature here --
72
213000
3000
근육에다가 걸어주고
03:48
and hope that it does something interesting.
73
216000
3000
뭔가 재미난 일이 벌어지기를 기대합니다.
03:51
At the beginning, they're all going to be very boring.
74
219000
2000
처음에는 정말 지루하죠
03:53
Most of them won't move at all,
75
221000
2000
대부분은 움직이지도 않고
03:55
but some of them might make a tiny step.
76
223000
2000
간혹 일부는 한발짝을 떼기도 합니다.
03:57
Those are then selected by the algorithm,
77
225000
2000
알고리즘에 따라 일부를 선택해서
03:59
reproduced with mutation and recombinations to introduce sex as well.
78
227000
4000
복제와 재조합을 통해 재생산하고 성별까지 부여합니다.
04:03
And you repeat that process over and over again,
79
231000
2000
그리고 다시 그 과정을 계속해서 반복하는데
04:05
until you have something that walks --
80
233000
2000
언제까지 하냐면
04:07
in this case, in a straight line, like this.
81
235000
2000
이렇게 똑바로 걷는 녀석이 나올때 까지 합니다.
04:09
So that was the idea behind this.
82
237000
2000
이게 바로 오늘 얘기할 내용입니다.
04:11
When we started this, I set up the simulation one evening.
83
239000
3000
처음 할때는 저녁에 시뮬레이션을 작동하고,
04:14
It took about three to four hours to run the simulation.
84
242000
3000
서너시간 걸리니까
04:17
I got up the next morning, went to the computer and looked at the results,
85
245000
4000
다음날 아침 일어나서 컴퓨터로 가서 결과를 봅니다.
04:21
and was hoping for something that walked in a straight line,
86
249000
3000
지금 방금 제가 걸은 것처럼 뭔가가
04:24
like I've just demonstrated,
87
252000
2000
똑바로 걸어가기를 기대하면서 말이죠
04:26
and this is what I got instead.
88
254000
2000
그런데 나온 것은 이렇습니다.
04:28
(Laughter)
89
256000
10000
웃음
04:38
So, it was back to the drawing board for us.
90
266000
3000
저희는 처음부터 다시 시작해서
04:42
We did get it to work eventually,
91
270000
3000
여기저기 좀 조정한 후에 결국
04:45
after tweaking a bit here and there.
92
273000
2000
걷도록 만들었습니다.
04:47
And this is an example of a successful evolutionary run.
93
275000
3000
그리고 이건 성공적으로 진화한 샘플입니다.
04:50
So, what you'll see in a moment is a very simple biped
94
278000
3000
잠시후 보실것은 인공진화과정을 통해 걷는 것을 배운
04:53
that's learning how to walk using artificial evolution.
95
281000
3000
두발달린 캐릭터 입니다.
04:56
At the beginning, it can't walk at all,
96
284000
2000
처음에는 전혀 걸을 수 없다가
04:58
but it will get better and better over time.
97
286000
2000
시간이 갈수록 점점 좋아지게 될 겁니다.
05:02
So, this is the one that can't walk at all.
98
290000
3000
이건 전혀 걸을 수 없는 거구요
05:05
(Laughter)
99
293000
6000
(웃음)
05:11
Now, after five generations of applying evolutionary process,
100
299000
3000
5세대가 진화하면
05:14
the genetic algorithm is getting a tiny bit better.
101
302000
3000
유전알고리즘은 아주 조금 향상됩니다.
05:17
(Laughter)
102
305000
8000
(웃음)
05:25
Generation 10 and it'll take a few steps more --
103
313000
2000
10세대가 되면 몇 발짝 걷지만
05:31
still not quite there.
104
319000
2000
아직 멀었죠
05:34
But now, after generation 20, it actually walks in a straight line without falling over.
105
322000
5000
하지만 20세대가 지나서는 넘어지지 않고 똑바로 걷게 됩니다.
05:40
That was the real breakthrough for us.
106
328000
3000
놀라운 발전이죠
05:43
It was, academically, quite a challenging project,
107
331000
3000
학문적으로도 매우 어려운 일이기도 했습니다.
05:46
and once we had reached that stage, we were quite confident
108
334000
3000
여기까지 발전하자 자신감이 들더군요
05:49
that we could try and do other things as well with this approach --
109
337000
3000
사람의 몸과, 신경회로를 흉내내는 이 방법을
05:52
actually simulating the body
110
340000
2000
활용하서
05:54
and simulating that part of the nervous system that controls it.
111
342000
3000
다른 일들도 할 수 있겠다는 생각이 들었습니다.
05:57
Now, at this stage, it also became clear that this could be very exciting
112
345000
3000
지금 이단계 만으로도 컴퓨터 게임이나 온라인 상에서의
06:00
for things like computer games or online worlds.
113
348000
3000
일들이 매우 재미있어 질 수 있습니다.
06:03
What you see here is the character standing there,
114
351000
2000
지금 보시는 것은 한 캐릭터가 서 있는데,
06:05
and there's an obstacle that we put in its way.
115
353000
2000
앞에다가 장애물을 놓으면
06:07
And what you see is, it's going to fall over the obstacle.
116
355000
5000
장애물에 걸려 넘어지게 됩니다.
06:12
Now, the interesting bit is, if I move the obstacle a tiny bit to the right,
117
360000
3000
재미있는 것은 장애물을 조금 우츨으로 옮기면
06:15
which is what I'm doing now, here,
118
363000
2000
이렇게 말이죠
06:17
it will fall over it in a completely different way.
119
365000
4000
전혀 다르게 넘어지는 것을 볼 수 있습니다.
06:24
And again, if you move the obstacle a tiny bit, it'll again fall differently.
120
372000
5000
그리고 다시 조금 움직이면, 역시 다르게 넘어집니다.
06:29
(Laughter)
121
377000
2000
(웃음)
06:31
Now, what you see, by the way, at the top there,
122
379000
2000
이번에 보실 여기 위에 있는 것은
06:33
are some of the neural activations being fed into the virtual muscles.
123
381000
3000
신경활성화기능이 있는 가상의 근육입니다.
06:36
Okay. That's the video. Thanks.
124
384000
2000
네 바로 이겁니다. 감사합니다.
06:38
Now, this might look kind of trivial, but it's actually very important
125
386000
3000
별거 없어보이지만 사실 매우 중요합니다.
06:41
because this is not something you get at the moment
126
389000
2000
왜냐하면 이제까지 어떠한 가상세계에서도
06:43
in any interactive or any virtual worlds.
127
391000
2000
존재하지 않았으니깐요
06:48
Now, at this stage, we decided to start a company and move this further,
128
396000
3000
이단계에서 우리는 회사를 세우고 좀더 연구하기로 했습니다.
06:51
because obviously this was just a very simple, blocky biped.
129
399000
3000
이건 아주 간단한 짤딸막한 두발달린 캐릭터에 불과하지만요
06:54
What we really wanted was a full human body.
130
402000
2000
우리가 원했던 것은 실제 사람같은 캐릭터 였기 때문에
06:56
So we started the company.
131
404000
1000
더 연구하기 위해 회사를 설립했습니다.
06:57
We hired a team of physicists, software engineers and biologists
132
405000
5000
물리학자, 소프트웨어 엔지니어, 생물학자로 구성된 팀을
07:02
to work on this, and the first thing we had to work on
133
410000
3000
고용하고 처음으로 한것이 바로
07:05
was to create the human body, basically.
134
413000
4000
인간의 몸을 흉내내는 것이었습니다.
07:09
It's got to be relatively fast, so you can run it on a normal machine,
135
417000
3000
이제 꽤 빨라서 일반 컴퓨터에서도 작동시킬 수 있는데
07:12
but it's got to be accurate enough, so it looks good enough, basically.
136
420000
3000
실제처럼 보일 정도로 정확해야 합니다.
07:15
So we put quite a bit of biomechanical knowledge into this thing,
137
423000
3000
그래서 우리는 상당한 수준의 생물역학적 지식도 쏟아 넣어
07:18
and tried to make it as realistic as possible.
138
426000
4000
가능한 실제처럼 보이도록 했습니다.
07:22
What you see here on the screen right now
139
430000
2000
지금 스크린에 보이는 것은
07:24
is a very simple visualization of that body.
140
432000
2000
그렇게 만든 사람의 간단한 영상입니다.
07:26
I should add that it's very simple to add things like hair, clothes, etc.,
141
434000
4000
머리카락이나 옷을 입히는 것은 간단하지만
07:30
but what we've done here is use a very simple visualization,
142
438000
3000
여기서는 여러분이 움직임에 집중할 수 있도록
07:33
so you can concentrate on the movement.
143
441000
2000
최대한 단순화 하였습니다.
07:35
Now, what I'm going to do right now, in a moment,
144
443000
3000
이제 제가 할 것은
07:38
is just push this character a tiny bit and we'll see what happens.
145
446000
3000
캐릭터를 조금 밀면, 어떤일이 벌어지는지 보시게 됩니다.
07:46
Nothing really interesting, basically.
146
454000
2000
별다른 것은 없죠
07:48
It falls over, but it falls over like a rag doll, basically.
147
456000
3000
천으로 만든 인형처럼 넘어지죠
07:51
The reason for that is that there's no intelligence in it.
148
459000
3000
그 이유는 아직 지능이 없어서 입니다.
07:54
It becomes interesting when you put artificial intelligence into it.
149
462000
4000
인공지능을 추가하면 재미있어지는데요
07:58
So, this character now has motor skills in the upper body --
150
466000
4000
이제 몸 윗쪽에 운동기능을 갖게 됩니다.
08:02
nothing in the legs yet, in this particular one.
151
470000
2000
아직 다리에는 없지만.
08:04
But what it will do -- I'm going to push it again.
152
472000
3000
제가 다시 밀면
08:07
It will realize autonomously that it's being pushed.
153
475000
2000
누군가 밀고있다는 것을 자동적으로 감지하고
08:09
It's going to stick out its hands.
154
477000
2000
손을 앞으로 뻗으면서
08:11
It's going to turn around into the fall, and try and catch the fall.
155
479000
3000
돌아서 넘어지려고 합니다.
08:20
So that's what you see here.
156
488000
2000
이렇게 말이죠
08:22
Now, it gets really interesting
157
490000
2000
이제 재미있어 지는데요
08:24
if you then add the AI for the lower part of the body as well.
158
492000
4000
인공지능을 다리에도 추가하면
08:28
So here, we've got the same character.
159
496000
2000
여기 동일한 캐릭터에가 있는데,
08:30
I'm going to push it a bit harder now,
160
498000
2000
조금 밀어 보겠습니다.
08:32
harder than I just pushed Chris.
161
500000
2000
Chris 밀었던 것 보다 좀더 세게
08:34
But what you'll see is -- it's going to receive a push now from the left.
162
502000
4000
그러면 캐릭터는 왼쪽에서 미는 것을 느끼고
08:41
What you see is it takes steps backwards,
163
509000
2000
뒷걸음 치면서
08:43
it tries to counter-balance,
164
511000
2000
균형을 잡으로고 하고
08:45
it tries to look at the place where it thinks it's going to land.
165
513000
4000
넘어질 장소를 보게되죠.
08:49
I'll show you this again.
166
517000
2000
다시 보여드리겠습니다.
08:51
And then, finally hits the floor.
167
519000
3000
그리고 결국은 바닥에 넘어지죠
08:55
Now, this becomes really exciting
168
523000
3000
이제부터 점점 재미있어 지는데요
08:58
when you push that character in different directions, again, just as I've done.
169
526000
5000
제가 했던것처럼 좀 다른 방향에서 밀게되면 재미있는 것을 볼 수 있습니다.
09:03
That's something that you cannot do right now.
170
531000
4000
이런것은 여러분은 지금 할 수 없는 것이죠
09:07
At the moment, you only have empty computer graphics in games.
171
535000
3000
지금 게임속의 캐릭터는 속빈 그래픽에 불과하기 때문입니다
09:10
What this is now is a real simulation. That's what I want to show you now.
172
538000
3000
이게 바로 제가 보여드리고자 하는 진정한 사람을 흉내낸 캐릭터 인데
09:13
So, here's the same character with the same behavior I've just shown you,
173
541000
3000
아까보신 것과 같은 행동을 합니다.
09:16
but now I'm just going to push it from different directions.
174
544000
2000
그런데 제가 다른 방향에서 밀어보겠습니다.
09:18
First, starting with a push from the right.
175
546000
2000
우선 우측에서 밀어 봅니다.
09:23
This is all slow motion, by the way, so we can see what's going on.
176
551000
3000
참고로 이건 느린동작이라 자세히 보실 수 있습니다.
09:26
Now, the angle will have changed a tiny bit,
177
554000
3000
방향을 조금 바꾸어 밀면
09:29
so you can see that the reaction is different.
178
557000
4000
반응이 다른 것을 보실 수 있습니다.
09:33
Again, a push, now this time from the front.
179
561000
3000
이번에는 앞에서 밀면
09:37
And you see it falls differently.
180
565000
2000
다르게 넘어지죠?
09:39
And now from the left --
181
567000
2000
좌측에서 밀면
09:43
and it falls differently.
182
571000
2000
역시 다르게 넘어집니다.
09:45
That was really exciting for us to see that.
183
573000
2000
이걸 처음 보았을때
09:47
That was the first time we've seen that.
184
575000
2000
아주 흥분되었습니다.
09:49
This is the first time the public sees this as well,
185
577000
2000
이 장면은 여태것 어디에도 보여준 적이 없는 것으로 여러분은
09:51
because we have been in stealth mode.
186
579000
2000
세상에서 처음으로
09:53
I haven't shown this to anybody yet.
187
581000
2000
지금 이 장면을 보고계십니다.
09:55
Now, just a fun thing:
188
583000
2000
이제 재미있는 일은
09:57
what happens if you put that character --
189
585000
2000
이 캐릭터에다가
09:59
this is now a wooden version of it, but it's got the same AI in it --
190
587000
2000
현대적인 인공지능을 추가하고
10:01
but if you put that character on a slippery surface, like ice.
191
589000
2000
얼음처럼 미끄러운 바닥에 올려 놓으면
10:03
We just did that for a laugh, just to see what happens.
192
591000
3000
장난삼아 해본건데 아주 재미있습니다.
10:06
(Laughter)
193
594000
1000
(웃음)
10:07
And this is what happens.
194
595000
2000
재미있죠?
10:09
(Laughter)
195
597000
3000
(웃음)
10:12
(Applause)
196
600000
3000
(박수)
10:15
It's nothing we had to do about this.
197
603000
2000
우리가 이런 동작을 만든것은 아닙니다.
10:17
We just took this character that I just talked about,
198
605000
2000
단지 설명드린 캐릭터를 만들어서
10:19
put it on a slippery surface, and this is what you get out of it.
199
607000
3000
미끄러운 바닥에 올려놓은 것 분인데 이렇게 움직입니다.
10:22
And that's a really fascinating thing about this approach.
200
610000
3000
아주 신나는 일이죠
10:26
Now, when we went to film studios and games developers
201
614000
3000
우리가 이 캐릭터를 가지고 영화 스튜디오와
10:29
and showed them that technology, we got a very good response.
202
617000
3000
게임개발자들을 찾아갔을때 반응은 매우 뜨거웠습니다.
10:32
And what they said was, the first thing they need immediately is virtual stuntmen.
203
620000
4000
당장 가상의 스턴트맨이 필요했다고 하더군요
10:36
Because stunts are obviously very dangerous, they're very expensive,
204
624000
4000
위험한 일을 하다보니 매우 비싼데다가
10:40
and there are a lot of stunt scenes that you cannot do, obviously,
205
628000
2000
스턴트맨이 할 수 없는 많은 장면들도 많이 있어
10:42
because you can't really allow the stuntman to be seriously hurt.
206
630000
3000
다치게 할수 는 없으니 고민이라더군요
10:45
So, they wanted to have a digital version of a stuntman
207
633000
3000
그래서 가상의 스턴트맨을 찾고 있었는데
10:48
and that's what we've been working on for the past few months.
208
636000
2000
그건 바로 우리가 몇달간 작업해온 것이며
10:50
And that's our first product that we're going to release in a couple of weeks.
209
638000
5000
몇주 후에 시장에 출시할 첫제품이기도 합니다.
10:55
So, here are just a few very simple scenes of the guy just being kicked.
210
643000
5000
여러분은 최근 완성된 몇 개의 간단한 장면들을 보고 계십니다.
11:00
That's what people want. That's what we're giving them.
211
648000
2000
이것이 바로 세상이 원하는 것이며 우리가 주려고 하는 것입니다.
11:02
(Laughter)
212
650000
7000
(웃음)
11:09
You can see, it's always reacting.
213
657000
2000
느끼시겠지만 상당히 현실적이죠
11:11
This is not a dead body. This is a body who basically, in this particular case,
214
659000
4000
딱딱한 캐릭터가 아니라, 보시다 시피
11:15
feels the force and tries to protect its head.
215
663000
2000
외부의 힘을 느끼고 이에 대응해서 머리를 보호하려고 합니다.
11:17
Only, I think it's quite a big blow again.
216
665000
2000
굉장한 힘이 가해집니다.
11:19
You feel kind of sorry for that thing,
217
667000
2000
여러분은 불쌍하게 느끼시겠지만,
11:21
and we've seen it so many times now that
218
669000
2000
우리는 너무 많이 봐서
11:23
we don't really care any more.
219
671000
2000
별로 걱정은 안됩니다.
11:25
(Laughter)
220
673000
1000
(웃음)
11:26
There are much worse videos than this, by the way, which I have taken out, but ...
221
674000
4000
이거보다 훨씬 더한 장면도 있지만 여기서는 제외하였습니다.
11:31
Now, here's another one.
222
679000
2000
한가지 더 보여드릴게 있는데
11:33
What people wanted as a behavior was to have an explosion,
223
681000
4000
사람들은 강력한 힘이 가해지는 폭발이 있을때
11:37
a strong force applied to the character,
224
685000
2000
캐릭터가 힘에 대해
11:39
and have the character react to it in midair.
225
687000
2000
어떻게 반응하는지를 보고싶어 했습니다.
11:41
So that you don't have a character that looks limp,
226
689000
2000
그래서 폭발이 있으면 축처지는 것이 아닌
11:43
but actually a character that you can use in an action film straight away,
227
691000
3000
실제 공중에서 살아있는 듯한 캐릭터를 활용해서
11:46
that looks kind of alive in midair as well.
228
694000
2000
바로 액션영화에 활용할수 있기를 원했습니다.
11:48
So this character is going to be hit by a force,
229
696000
2000
여기 이 캐릭터는 어떤 힘에 의해 날아갑니다.
11:50
it's going to realize it's in the air,
230
698000
2000
몸이 공중에 떠있는 것을 느끼게 되고
11:52
and it's going to try and, well,
231
700000
3000
떨어지는 방향으로
11:55
stick out its arm in the direction where it's landing.
232
703000
2000
팔을 뻗게 됩니다.
11:59
That's one angle; here's another angle.
233
707000
3000
이번에는 다른 각도입니다.
12:02
We now think that the realism we're achieving with this
234
710000
2000
이 캐릭터를 통해 실현되는 현실감은
12:04
is good enough to be used in films.
235
712000
2000
충분히 영화에 적용될 수 있을 듯 합니다.
12:06
And let's just have a look at a slightly different visualization.
236
714000
3000
좀 다른 영상을 한번 보시죠
12:09
This is something I just got last night
237
717000
2000
이건 제가 바로 어제
12:11
from an animation studio in London, who are using our software
238
719000
3000
런던에 있는 스튜디오에서 우리가 만든 소프트웨어를
12:14
and experimenting with it right now.
239
722000
2000
활용해서 만든 겁니다.
12:16
So this is exactly the same behavior that you saw,
240
724000
3000
방금 보신것과 동일한 것이지만
12:19
but in a slightly better rendered version.
241
727000
4000
좀더 향상된 것입니다.
12:23
So if you look at the character carefully,
242
731000
3000
캐릭터를 자세히 보시면
12:26
you see there are lots of body movements going on,
243
734000
2000
무수히 많은 동작들이 포함된 것을 확인하실 수 있는데
12:28
none of which you have to animate like in the old days.
244
736000
2000
예전에는 이런동작이 불가능했습니다.
12:30
Animators had to actually animate them.
245
738000
2000
제작자는 실제로 사람을 흉내내야 합니다.
12:32
This is all happening automatically in the simulation.
246
740000
2000
시뮬레이션에서 자동적으로 일어나는 것들이죠.
12:34
This is a slightly different angle,
247
742000
2000
이번에는 좀 다른 각도에서의 모습인데
12:39
and again a slow motion version of this.
248
747000
2000
느린동작으로 보겠습니다.
12:41
This is incredibly quick. This is happening in real time.
249
749000
4000
정말 빠르죠? 실시간으로 일어나는 것입니다
12:45
You can run this simulation in real time, in front of your eyes,
250
753000
2000
눈앞에서 실제와 같은 장면을 시뮬레이션 하고,
12:47
change it, if you want to, and you get the animation straight out of it.
251
755000
3000
바꾸고 싶은게 있다면 바로 원하는 영상을 얻을 수 있습니다.
12:50
At the moment, doing something like this by hand
252
758000
2000
현재 이걸 수작업으로 한다면
12:52
would take you probably a couple of days.
253
760000
2000
아마 며칠은 족히 걸릴 겁니다.
12:55
This is another behavior they requested.
254
763000
3000
이 행동도 구현하고 싶다고 하더군요
12:58
I'm not quite sure why, but we've done it anyway.
255
766000
2000
왜 필요한지는 모르지만 한번 해보았습니다.
13:00
It's a very simple behavior that shows you the power of this approach.
256
768000
2000
이 접근방법이 얼마나 강력한지 알수 있는 간단한 행동을 합니다.
13:02
In this case, the character's hands
257
770000
2000
여기서 캐릭터의 손은
13:04
are fixed to a particular point in space,
258
772000
2000
공중 어딘가에 고정되어 있고
13:06
and all we've told the character to do is to struggle.
259
774000
3000
발버둥 처보라고 명령했더니
13:09
And it looks organic. It looks realistic.
260
777000
3000
마치 살아있는 사람같습니다.
13:12
You feel kind of sorry for the guy.
261
780000
2000
불쌍해 보이시나요?
13:14
It's even worse -- and that is another video I just got last night --
262
782000
3000
이건 제가 어제 작업한 또 다른 영상인데 좀 더 현실적으로
13:17
if you render that a bit more realistically.
263
785000
2000
만들면 더 불쌍해 보입니다.
13:23
Now, I'm showing this to you just to show you
264
791000
2000
제가 이걸 보여드리는 이유는
13:25
how organic it actually can feel, how realistic it can look.
265
793000
2000
얼마나 살아있고 현실감 있어 보이는지 보여드리기 위해서입니다.
13:27
And this is all a physical simulation of the body,
266
795000
3000
이건 사람의 몸을 실제와 같이 흉내내기 입니다.
13:30
using AI to drive virtual muscles in that body.
267
798000
3000
인공지능이 가상의 근육을 움직이게 돼죠.
13:35
Now, one thing which we did for a laugh was
268
803000
3000
장난삼아 만든 좀더 복잡한
13:38
to create a slightly more complex stunt scene,
269
806000
2000
스턴트 장면을 보여드리겠습니다.
13:40
and one of the most famous stunts is the one where James Bond
270
808000
3000
유명한 스턴트 장면인데 제임스 본드가 스위스에 있는
13:43
jumps off a dam in Switzerland and then is caught by a bungee.
271
811000
4000
댐에서 뛰어내리고 마지막에 번지줄에 걸립니다.
13:48
Got a very short clip here.
272
816000
2000
여기있군요
13:54
Yes, you can just about see it here.
273
822000
2000
한번 감상해 보시죠.
13:56
In this case, they were using a real stunt man. It was a very dangerous stunt.
274
824000
3000
여기서는 실제 스턴트맨이 점프했는데 정말 위험한 일이죠
13:59
It was just voted, I think in the Sunday Times, as one of the most impressive stunts.
275
827000
3000
아마 Sunday Times 였던거 같은데요 가장 기억에 남는 장면으로 뽑혔다고 하네요
14:02
Now, we've just tried and -- looked at our character and asked ourselves,
276
830000
3000
이장면을 우리가 만든 캐릭터를 통해 만들어 보기로 했습니다.
14:05
"Can we do that ourselves as well?"
277
833000
2000
과연 할 수 있을까 궁금했습니다.
14:07
Can we use the physical simulation of the character,
278
835000
2000
캐릭터 신체 시뮬레이션을 위해
14:09
use artificial intelligence,
279
837000
2000
인공지능을
14:11
put that artificial intelligence into the character,
280
839000
2000
캐릭터에 추가하고
14:13
drive virtual muscles, simulate the way he jumps off the dam,
281
841000
4000
가상의 근육도 추가해서 본드가 댐에서 점프하고
14:17
and then skydive afterwards,
282
845000
2000
공중에서 다이빙해서 내려오다가
14:19
and have him caught by a bungee afterwards?
283
847000
2000
번지줄에 걸리는 장면을 흉내낼 수 있을까?
14:21
We did that. It took about altogether just two hours,
284
849000
3000
2시간 정도 걸려서
14:24
pretty much, to create the simulation.
285
852000
2000
그 장면을 흉내내어 보았습니다.
14:26
And that's what it looks like, here.
286
854000
2000
그랬더니 이런 결과가 나왔습니다.
14:37
Now, this could do with a bit more work. It's still very early stages,
287
865000
3000
좀더 작업을 할 수 있지만, 이건 아주 초기단계라서
14:40
and we pretty much just did this for a laugh,
288
868000
2000
장난삼아서 우리가 무엇을 얻어낼수 있는지
14:42
just to see what we'd get out of it.
289
870000
2000
만들어 본 것입니다.
14:44
But what we found over the past few months
290
872000
2000
그런데 우리가 지난 몇달간 발견한 것은
14:46
is that this approach -- that we're pretty much standard upon --
291
874000
3000
상당히 표준화된 이방법이 놀라울 정도로
14:49
is incredibly powerful.
292
877000
2000
강력하다는 것입니다.
14:51
We are ourselves surprised what you actually get out of the simulations.
293
879000
4000
우리들 자신도 시뮬레이션 결과를 보고 놀랐습니다.
14:55
There's very often very surprising behavior that you didn't predict before.
294
883000
4000
예전에는 상상할 수 없었던 놀라운 행동들이 가능합니다.
14:59
There's so many things we can do with this right now.
295
887000
2000
이걸 가지고 아주 많은 일들을 당장 할 수 있습니다.
15:01
The first thing, as I said, is going to be virtual stuntmen.
296
889000
3000
우선은 가상의 스턴트맨을 만드는 겁니다.
15:04
Several studios are using this software now to produce virtual stuntmen,
297
892000
4000
여러 스튜디오에서 가상의 스턴트맨을 만들기 위해 우리의 소프트웨어를 사용하고 있으며
15:08
and they're going to hit the screen quite soon, actually,
298
896000
2000
조만간 몇몇 대형 제작자들이 이를 실제 영화에 활용하여
15:10
for some major productions.
299
898000
2000
방영할 예정입니다.
15:12
The second thing is video games.
300
900000
3000
두번째로 활용가능한 분야는 비디오 게임입니다.
15:15
With this technology, video games will look different and they will feel very different.
301
903000
4000
이 기술은 비디오게임을 다르게 보이고 느껴지도록 만듭니다.
15:19
For the first time, you'll have actors that really feel very interactive,
302
907000
3000
여러분은 이제 처음으로 실제와 같이 반응하고
15:22
that have real bodies that really react.
303
910000
2000
실제 육체를 가진 캐릭터를 가지게 될 것입니다.
15:24
I think that's going to be incredibly exciting.
304
912000
3000
정말 대단한 일이죠?
15:27
Probably starting with sports games,
305
915000
2000
아마 스포츠 게임처럼 상호작용이 많은
15:29
which are going to become much more interactive.
306
917000
2000
부분부터 시작할 것입니다.
15:31
But I particularly am really excited
307
919000
1000
제가 특히 온라인에서 이 기술을 사용할 경우
15:32
about using this technology in online worlds,
308
920000
3000
흥분되는 것은, 예를 들어,
15:35
like there, for example, that Tom Melcher has shown us.
309
923000
3000
Tom Melcher가 우리에게 보여준 것 입니다.
15:38
The degree of interactivity you're going to get
310
926000
2000
상호작용의 정도는 여지까지의 것과는
15:40
is totally different, I think, from what you're getting right now.
311
928000
3000
완전히 다른 수준입니다.
15:44
A third thing we are looking at and very interested in is simulation.
312
932000
4000
세번째로 우리가 보고있는 것은 아주흥미롭습니다.
15:49
We've been approached by several simulation companies,
313
937000
2000
여러 시뮬레이션 회사들이 제안을 해왔지만
15:51
but one project we're particularly excited about, which we're starting next month,
314
939000
3000
특히 관심을 끄는 한가지가 있는데, 바로 다음달 부터 시작합니다.
15:54
is to use our technology -- and in particular, the walking technology --
315
942000
4000
그건 우리의 기술 그 중에서도 특히 걷는 기술을
15:58
to help aid surgeons who work on children with cerebral palsy,
316
946000
4000
외과의사들이 뇌성마비가 있는 아이들의 수술 결과가
16:02
to predict the outcome of operations on these children.
317
950000
3000
어떨지 예측할 수 있도록 도와주데 사용하는 것입니다.
16:05
As you probably know,
318
953000
2000
아마 아시겠지만
16:07
it's very difficult to predict what the outcome of an operation is
319
955000
3000
걸음걸이가 어떻게 바뀔지 수술 결과를 예측한다는 것은 정말이지
16:10
if you try and correct the gait.
320
958000
2000
어려운 일입니다.
16:12
The classic quote is, I think, it's unpredictable at best,
321
960000
3000
열길 물속은 알아도 한길 사람속은 모른다는 속담이 있지만, 이 결과 역시
16:15
is what people think right now, is the outcome.
322
963000
3000
아무도 알 수 없습니다.
16:18
Now, what we want to do with our software is allow our surgeons to have a tool.
323
966000
4000
우리의 소프트웨어가 외과의사들에게 좋은 수단이 되었으면 합니다.
16:22
We're going to simulate the gait of a particular child
324
970000
3000
우리는 한 아이의 걸음걸이를 흉내내고,
16:25
and the surgeon can then work on that simulation
325
973000
3000
의사선생님도 참여해서
16:28
and try out different ways to improve that gait,
326
976000
2000
걸음걸이를 교정하기 위해 여러 방법들을 시도한 후에
16:30
before he actually commits to an actual surgery.
327
978000
3000
의사선생님이 실제 수술에 들어갈겁니다.
16:33
That's one project we're particularly excited about,
328
981000
2000
이건 정말 흥분되는 일인데요
16:35
and that's going to start next month.
329
983000
2000
바로 다음달에 시작하려합니다.
16:39
Just finally, this is only just the beginning.
330
987000
3000
마지막으로, 여기까지는 시작에 불과합니다.
16:42
We can only do several behaviors right now.
331
990000
2000
여러가지 것들을 당장이라도 할 수 있죠.
16:44
The AI isn't good enough to simulate a full human body.
332
992000
3000
인공지능만으로 사람을 흉내내는 것은 어렵습니다.
16:47
The body yes, but not all the motor skills that we have.
333
995000
3000
특히 모든 운동기능을 구현하기는 더욱 어렵죠.
16:50
And, I think, we're only there if we can have something like ballet dancing.
334
998000
3000
발레춤추는 것을 흉내내려면 모든 운동기능을 구현해야하는데
16:53
Right now, we don't have that
335
1001000
2000
아직 거기까지는 아닙니다.
16:55
but I'm very sure that we will be able to do that at some stage.
336
1003000
2000
하지만 언젠가는 가능하리라 봅니다.
16:57
We do have one unintentional dancer actually,
337
1005000
3000
마지막으로 보여드릴 것은 춤추는 사람인데요
17:00
the last thing I'm going to show you.
338
1008000
2000
의도한건 아닙니다.
17:02
This was an AI contour that was produced and evolved --
339
1010000
3000
이건 인공지능을 갖춘 모습인데
17:05
half-evolved, I should say -- to produce balance, basically.
340
1013000
3000
균형감을 갖추기 위해 반쯤 진화된 수준입니다.
17:08
So, you kick the guy and the guy's supposed to counter-balance.
341
1016000
3000
이사람을 걷어차면 균형을 잡도록 되어 있죠.
17:11
That's what we thought was going to come out of this.
342
1019000
3000
그게 우리가 원하던 모습인데
17:14
But this is what emerged out of it, in the end.
343
1022000
2000
실제는 이렇습니다.
17:17
(Music)
344
1025000
10000
(음악)
17:27
Bizarrely, this thing doesn't have a head. I'm not quite sure why.
345
1035000
3000
이상하게도 머리가 없는데 왜그런지는 저도 잘 모릅니다.
17:31
So, this was not something we actually put in there.
346
1039000
2000
우리가 그렇게 만든건 아니란 얘기죠.
17:33
He just started to create that dance himself.
347
1041000
4000
스스로 춤을 추는데
17:37
He's actually a better dancer than I am, I have to say.
348
1045000
3000
사실 저보다 잘추는 군요.
17:41
And what you see after a while --
349
1049000
2000
잠시뒤에는
17:43
I think he even goes into a climax right at the end.
350
1051000
2000
절정의 마지막을 장식하게 됩니다.
17:49
And I think -- there you go.
351
1057000
3000
보이시죠?
17:52
(Laughter)
352
1060000
2000
(웃음)
17:54
So, that all happened automatically. We didn't put that in there.
353
1062000
2000
모든게 자동으로 나오는 영상으로 우리가 한건 없습니다.
17:56
That's just the simulation creating this itself, basically.
354
1064000
3000
기본적으로 스스로를 창조하는 시뮬레이션 이죠.
17:59
So it's just --
355
1067000
2000
아직은
18:01
(Applause)
356
1069000
1000
(박수)
18:02
Thanks.
357
1070000
2000
감사합니다.
18:05
Not quite John Travolta yet, but we're working on that as well,
358
1073000
3000
아직은 존트라볼타까지는 아니지만 이미 작업 중이랍니다.
18:08
so thanks very much for your time.
359
1076000
2000
시간 내주셔서 감사드립니다.
18:10
Thanks.
360
1078000
1000
감사합니다.
18:11
(Applause)
361
1079000
1000
박수
18:12
CA: Incredible. That was really incredible.
362
1080000
2000
Chris Anderson: 정말 놀랍군요.
18:14
TR: Thanks.
363
1082000
1000
Torsten Reil:감사합니다.
ABOUT THE SPEAKER
Torsten Reil - Animating neurobiologistBy coding computer simulations with biologically modeled nervous systems, Torsten Reil and his company NaturalMotion breathe life into the animated characters inhabiting the most eye-poppingly realistic games and movies around.
Why you should listen
From modeling the mayhem of equine combat in Lord of the Rings: Return of the King to animating Liberty City gun battles in Grand Theft Auto IV, Torsten Reil's achievements are all over the map these days. Software that he helped create (with NaturalMotion, the imaging company he co-founded) has revolutionized computer animation of human and animal avatars, giving rise to some of the most breathtakingly real sequences in the virtual world of video games and movies- and along the way given valuable insight into the way human beings move their bodies.
Reil was a neural researcher working on his Masters at Oxford, developing computer simulations of nervous systems based on genetic algorithms- programs that actually used natural selection to evolve their own means of locomotion. It didn't take long until he realized the commercial potential of these lifelike characters. In 2001 he capitalized on this lucrative adjunct to his research, and cofounded NaturalMotion. Since then the company has produced motion simulation programs like Euphoria and Morpheme, state of the art packages designed to drastically cut the time and expense of game development, and create animated worlds as real as the one outside your front door. Animation and special effects created with Endorphin (NaturalMotion's first animation toolkit) have lent explosive action to films such as Troy and Poseidon, and NaturalMotion's software is also being used by LucasArts in video games such as the hotly anticipated Indiana Jones.
But there are serious applications aside from the big screen and the XBox console: NaturalMotion has also worked under a grant from the British government to study the motion of a cerebral palsy patient, in hopes of finding therapies and surgeries that dovetail with the way her nervous system is functioning.
More profile about the speakerReil was a neural researcher working on his Masters at Oxford, developing computer simulations of nervous systems based on genetic algorithms- programs that actually used natural selection to evolve their own means of locomotion. It didn't take long until he realized the commercial potential of these lifelike characters. In 2001 he capitalized on this lucrative adjunct to his research, and cofounded NaturalMotion. Since then the company has produced motion simulation programs like Euphoria and Morpheme, state of the art packages designed to drastically cut the time and expense of game development, and create animated worlds as real as the one outside your front door. Animation and special effects created with Endorphin (NaturalMotion's first animation toolkit) have lent explosive action to films such as Troy and Poseidon, and NaturalMotion's software is also being used by LucasArts in video games such as the hotly anticipated Indiana Jones.
But there are serious applications aside from the big screen and the XBox console: NaturalMotion has also worked under a grant from the British government to study the motion of a cerebral palsy patient, in hopes of finding therapies and surgeries that dovetail with the way her nervous system is functioning.
Torsten Reil | Speaker | TED.com