Abe Davis: New video technology that reveals an object's hidden properties

204,153 views ใƒป 2015-05-05

TED


์•„๋ž˜ ์˜๋ฌธ์ž๋ง‰์„ ๋”๋ธ”ํด๋ฆญํ•˜์‹œ๋ฉด ์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค.

๋ฒˆ์—ญ: Min Lee ๊ฒ€ํ† : Sungho Yoo
00:13
Most of us think of motion as a very visual thing.
0
13373
3349
์‚ฌ๋žŒ๋“ค์€ ๋ณดํ†ต ์›€์ง์ž„์„ ์‹œ๊ฐ์ ์ธ ๊ฒƒ์œผ๋กœ ์ƒ๊ฐํ•ฉ๋‹ˆ๋‹ค.
00:17
If I walk across this stage or gesture with my hands while I speak,
1
17889
5088
๋งŒ์•ฝ ์ œ๊ฐ€ ์ง€๊ธˆ ๋งํ•˜๋Š” ๋„์ค‘ ๋ฌด๋Œ€๋ฅผ ๊ฐ€๋กœ์งˆ๋Ÿฌ ๊ฑท๊ฑฐ๋‚˜ ์†๋™์ž‘์„ ์ทจํ•˜๋ฉด
00:22
that motion is something that you can see.
2
22977
2261
์—ฌ๋Ÿฌ๋ถ„์€ ๊ทธ ์›€์ง์ž„์„ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
00:26
But there's a world of important motion that's too subtle for the human eye,
3
26255
5482
ํ•˜์ง€๋งŒ ์ค‘์š”ํ•œ ์˜๋ฏธ๋ฅผ ๊ฐ–๋Š” ์›€์ง์ž„๋“ค ์ค‘
๊ทธ ํฌ๊ธฐ๊ฐ€ ๋งค์šฐ ๋ฏธ์„ธํ•ด์„œ ์šฐ๋ฆฌ ๋ˆˆ์— ๋ถ„๋ณ„ํ•˜๊ธฐ ์–ด๋ ค์šด ๊ฒƒ๋“ค๋„ ์žˆ์Šต๋‹ˆ๋‹ค.
00:31
and over the past few years,
4
31737
2041
์šฐ๋ฆฌ๋Š” ์ง€๋‚œ ๋ช‡ ๋…„์— ๊ฑธ์ณ ์ธ๊ฐ„์ด ๋ณด์ง€ ๋ชปํ•˜๋Š” ์ด ์›€์ง์ž„์„
00:33
we've started to find that cameras
5
33778
1997
00:35
can often see this motion even when humans can't.
6
35775
3410
์นด๋ฉ”๋ผ๋กœ๋Š” ์‹๋ณ„ ํ•  ์ˆ˜ ์žˆ๋‹ค๋Š” ์‚ฌ์‹ค์„ ์•Œ์•„๋ƒˆ์Šต๋‹ˆ๋‹ค.
00:40
So let me show you what I mean.
7
40305
1551
์–ด๋–ค ๊ฒƒ์ธ์ง€ ์ง์ ‘ ๋ณด์—ฌ๋“œ๋ฆฌ์ฃ .
00:42
On the left here, you see video of a person's wrist,
8
42717
3622
์™ผ์ชฝ ํ™”๋ฉด์€ ์‚ฌ๋žŒ ์†๋ชฉ์˜ ๋™์˜์ƒ์ž…๋‹ˆ๋‹ค.
00:46
and on the right, you see video of a sleeping infant,
9
46339
3147
์˜ค๋ฅธํŽธ์—๋Š” ์ž ์ž๋Š” ์•„๊ธฐ ๋™์˜์ƒ์ด ์žˆ์–ด์š”.
00:49
but if I didn't tell you that these were videos,
10
49486
3146
ํ•˜์ง€๋งŒ ์ด๊ฒƒ๋“ค์ด ๋™์˜์ƒ์ด๋ผ๊ณ  ๋ง์”€๋“œ๋ฆฌ์ง€ ์•Š์•˜๋‹ค๋ฉด
00:52
you might assume that you were looking at two regular images,
11
52632
3761
์—ฌ๋Ÿฌ๋ถ„๋“ค์€ ๋‘ ๊ฐœ์˜ ๋ณดํ†ต ์‚ฌ์ง„์„ ๋ณด๊ณ  ์žˆ๋‹ค๊ณ  ์ƒ๊ฐํ•˜์‹ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
00:56
because in both cases,
12
56393
1672
์™œ๋ƒํ•˜๋ฉด ๋‘ ์˜์ƒ ๋ชจ๋‘
00:58
these videos appear to be almost completely still.
13
58065
3047
์ •์ง€ํ•ด ์žˆ๋Š” ๊ฒƒ์ฒ˜๋Ÿผ ๋ณด์ด๊ฑฐ๋“ ์š”.
01:02
But there's actually a lot of subtle motion going on here,
14
62175
3885
ํ•˜์ง€๋งŒ ์—ฌ๊ธฐ์—๋Š” ์•„์ฃผ ๋ฏธ์„ธํ•œ ์›€์ง์ž„์ด ์ผ์–ด๋‚˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
01:06
and if you were to touch the wrist on the left,
15
66060
2392
๋งŒ์•ฝ ์—ฌ๋Ÿฌ๋ถ„์ด ์™ผ์ชฝ ์‚ฌ์ง„์˜ ์†๋ชฉ์„ ๋งŒ์งˆ ์ˆ˜ ์žˆ๋‹ค๋ฉด
01:08
you would feel a pulse,
16
68452
1996
๋งฅ๋ฐ•์„ ๋Š๋‚„ ์ˆ˜ ์žˆ์„ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
01:10
and if you were to hold the infant on the right,
17
70448
2485
๋˜ ๋งŒ์•ฝ ์˜ค๋ฅธ์ชฝ์˜ ์•„์ด๋ฅผ ์•ˆ์•„ ๋ณธ๋‹ค๋ฉด
01:12
you would feel the rise and fall of her chest
18
72933
2391
์•„์ด๊ฐ€ ์ˆจ์„ ์‰ด ๋•Œ๋งˆ๋‹ค ๊ฐ€์Šด์ด ์˜ค๋ฅด๋ฝ ๋‚ด๋ฆฌ๋ฝ
01:15
as she took each breath.
19
75324
1390
ํ•˜๋Š” ๊ฒƒ์„ ๋Š๋‚„ ์ˆ˜ ์žˆ์„ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
01:17
And these motions carry a lot of significance,
20
77762
3576
๋ณด์‹œ๋Š” ๊ฒƒ๊ณผ ๊ฐ™์€ ์›€์ง์ž„๋“ค์€ ์šฐ๋ฆฌ์—๊ฒŒ ์ค‘์š”ํ•œ ์ •๋ณด๋ฅผ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค.
01:21
but they're usually too subtle for us to see,
21
81338
3343
ํ•˜์ง€๋งŒ ๋Œ€๋ถ€๋ถ„์€ ์œก์•ˆ์œผ๋กœ ์‹๋ณ„ํ•˜๊ธฐ์— ๋งค์šฐ ๋ฏธ์„ธํ•ฉ๋‹ˆ๋‹ค.
01:24
so instead, we have to observe them
22
84681
2276
๊ทธ๋ž˜์„œ ์œก์•ˆ์œผ๋กœ ๊ด€์ฐฐํ•˜๋Š” ๋Œ€์‹ 
01:26
through direct contact, through touch.
23
86957
2900
์ด‰๊ฐ์„ ํ†ตํ•ด ์ง์ ‘ ๋งŒ์ ธ๋ณด๊ณ  ๋Š๋‚๋‹ˆ๋‹ค.
01:30
But a few years ago,
24
90997
1265
ํ•˜์ง€๋งŒ ๋ช‡ ๋…„ ์ „
01:32
my colleagues at MIT developed what they call a motion microscope,
25
92262
4405
MIT์˜ ์ œ ๋™๋ฃŒ๋“ค์ด "๋ชจ์…˜ ๋งˆ์ดํฌ๋กœ ์Šค์ฝฅ" ์ด๋ผ๋Š” ๊ฑธ ๊ฐœ๋ฐœํ–ˆ์Šต๋‹ˆ๋‹ค.
01:36
which is software that finds these subtle motions in video
26
96667
4384
์ด ์†Œํ”„ํŠธ์›จ์–ด๋Š” ์˜์ƒ์˜ ๋ฏธ์„ธํ•œ ์›€์ง์ž„์„ ์ฐพ์•„๋‚ด
01:41
and amplifies them so that they become large enough for us to see.
27
101051
3562
์šฐ๋ฆฌ๊ฐ€ ๋ณผ ์ˆ˜ ์žˆ์„ ๋งŒํผ์˜ ํฌ๊ธฐ๋กœ ๊ทธ ์›€์ง์ž„์„ ์ฆํญ์‹œ์ผœ์ค๋‹ˆ๋‹ค.
01:45
And so, if we use their software on the left video,
28
105416
3483
์™ผ์ชฝ ์˜์ƒ์— ์ด ์†Œํ”„ํŠธ์›จ์–ด๋ฅผ ์ ์šฉํ•ด ๋ณด๋ฉด
01:48
it lets us see the pulse in this wrist,
29
108899
3250
์†๋ชฉ์˜ ๋งฅ๋ฐ•์„ ๋ˆˆ์œผ๋กœ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
01:52
and if we were to count that pulse,
30
112149
1695
์šฐ๋ฆฌ๊ฐ€ ๋งŒ์•ฝ ์ด ๋งฅ๋ฐ•์„ ์„ผ๋‹ค๋ฉด
01:53
we could even figure out this person's heart rate.
31
113844
2355
์ด ์‚ฌ๋žŒ์˜ ์‹ฌ๋ฐ•์ˆ˜๋ฅผ ์•Œ์•„๋‚ผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
01:57
And if we used the same software on the right video,
32
117095
3065
๊ฐ™์€ ์†Œํ”„ํŠธ์›จ์–ด๋ฅผ ์˜ค๋ฅธ์ชฝ ์˜์ƒ์— ์‚ฌ์šฉํ•ด ๋ณด์•˜์Šต๋‹ˆ๋‹ค.
02:00
it lets us see each breath that this infant takes,
33
120160
3227
์•„์ด๊ฐ€ ์ˆจ์‰ฌ๋Š” ๊ฒƒ์„ ๋ˆˆ์œผ๋กœ ๋ณผ ์ˆ˜ ์žˆ๊ฒŒ ํ•ด์ฃผ๊ณ ,
02:03
and we can use this as a contact-free way to monitor her breathing.
34
123387
4137
์ง์ ‘ ๋งŒ์ง€์ง€ ์•Š๊ณ ๋„ ์•„๊ธฐ์˜ ํ˜ธํก์„ ๊ด€์ฐฐํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
02:08
And so this technology is really powerful because it takes these phenomena
35
128884
5348
์ด๋Ÿฌํ•œ ๊ธฐ์ˆ ์€ ๊ทธ ์˜ํ–ฅ๋ ฅ์ด ๋งค์šฐ ์—„์ฒญ๋‚ฉ๋‹ˆ๋‹ค.
๋ณดํ†ต ์šฐ๋ฆฌ๊ฐ€ ์ด‰๊ฐ์„ ํ†ตํ•ด ๋Š๋ผ๋Š” ํ˜„์ƒ๋“ค์„
02:14
that we normally have to experience through touch
36
134232
2367
์šฐํšŒ์ ์ธ ๋ฐฉ๋ฒ•์ธ ์‹œ๊ฐ์ ์œผ๋กœ ๊ฒฝํ—˜ํ•˜๊ฒŒ ํ•ด์ฃผ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
02:16
and it lets us capture them visually and non-invasively.
37
136599
2957
๋ช‡ ๋…„์ „ ์ €๋Š” ์ด ์†Œํ”„ํŠธ์›จ์–ด ๊ฐœ๋ฐœ์ž๋“ค๊ณผ ํ•จ๊ป˜ ์ผํ•˜๊ธฐ ์‹œ์ž‘ํ–ˆ์Šต๋‹ˆ๋‹ค.
02:21
So a couple years ago, I started working with the folks that created that software,
38
141104
4411
02:25
and we decided to pursue a crazy idea.
39
145515
3367
๊ทธ๋ฆฌ๊ณ  ๋ฏธ์นœ ๊ฒƒ์ฒ˜๋Ÿผ ๋“ค๋ฆฌ๋Š” ์ผ์„ ์‹œ๋„ํ•ด ๋ณด๊ธฐ๋กœ ํ–ˆ์Šต๋‹ˆ๋‹ค.
02:28
We thought, it's cool that we can use software
40
148882
2693
"์šฐ๋ฆฌ๊ฐ€ ์ƒ๊ฐํ•˜๊ธฐ์—,
์ด ์†Œํ”„ํŠธ์›จ์–ด๋ฅผ ํ†ตํ•ด ๋ฏธ๋™์„ ์‹œ๊ฐํ™”ํ•˜๋Š” ๊ฒƒ์€ ๋งค์šฐ ๋ฉ‹์ง„ ์ผ์ด๋ฉฐ
02:31
to visualize tiny motions like this,
41
151575
3135
02:34
and you can almost think of it as a way to extend our sense of touch.
42
154710
4458
์ด๋Š” ๋‹ค์‹œ๋งํ•ด ์ด‰๊ฐ์˜ ์—ญํ• ์„ ์—ฐ์žฅํ•œ ๊ฒƒ์ด๋ผ ํ•  ์ˆ˜ ์žˆ์ง€ ์•Š์€๊ฐ€.
02:39
But what if we could do the same thing with our ability to hear?
43
159168
4059
๊ทธ๋ ‡๋‹ค๋ฉด ์ด์™€ ๊ฐ™์€ ๊ธฐ์ˆ ์„ ํ†ตํ•ด ๋“ฃ๋Š” ๋Šฅ๋ ฅ์„ ํ™•์žฅํ•ด๋ณด๋ฉด ์–ด๋–จ๊นŒ?"
02:44
What if we could use video to capture the vibrations of sound,
44
164508
4665
๋งŒ์•ฝ ์ด ์˜์ƒ ์ฆํญ ๊ธฐ์ˆ ์„ ์ด์šฉํ•ด ์†Œ๋ฆฌ์˜ ์ง„๋™์„ ์ดฌ์˜ ํ•  ์ˆ˜ ์žˆ๋‹ค๋ฉด์š”?
02:49
which are just another kind of motion,
45
169173
2827
์†Œ๋ฆฌ์˜ ์ง„๋™ ์—ญ์‹œ ์•„์ฃผ ๋ฏธ์„ธํ•œ ์›€์ง์ž„์ด๋ผ๊ณ  ํ•  ์ˆ˜ ์žˆ์ž–์•„์š”.
02:52
and turn everything that we see into a microphone?
46
172000
3346
๊ทธ๋ฆฌ๊ณ  ์šฐ๋ฆฌ๊ฐ€ ๋ˆˆ์œผ๋กœ ๋ณผ ์ˆ˜ ์žˆ๋Š” ๊ฒƒ์„ ์†Œ๋ฆฌ๋กœ ๋ณ€ํ™˜ํ•ด ๋ณธ๋‹ค๋ฉด์š”?
02:56
Now, this is a bit of a strange idea,
47
176236
1971
๋‹ค์†Œ ์ด์ƒํ•œ ์ƒ๊ฐ์ฒ˜๋Ÿผ ๋ณด์ด๊ฒ ๋„ค์š”.
02:58
so let me try to put it in perspective for you.
48
178207
2586
์—ฌ๋Ÿฌ๋ถ„์„ ์œ„ํ•ด ๋‹ค๋ฅธ ๊ด€์ ์œผ๋กœ ์„ค๋ช…ํ•ด ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
03:01
Traditional microphones work by converting the motion
49
181523
3488
์ผ๋ฐ˜์ ์ธ ๋งˆ์ดํฌ์˜ ์›๋ฆฌ๋Š”
๋‚ด๋ถ€์˜ ์ง„๋™ํŒ์˜ ์›€์ง์ž„์„ ์ „๊ธฐ์‹ ํ˜ธ๋กœ ๋ณ€ํ™˜ํ•˜๋„๋ก ๋˜์–ด์žˆ๋Š”๋ฐ
03:05
of an internal diaphragm into an electrical signal,
50
185011
3599
03:08
and that diaphragm is designed to move readily with sound
51
188610
4318
์ง„๋™ํŒ์€ ์†Œ๋ฆฌ์— ๋ฐ˜์‘ํ•˜์—ฌ ์›€์ง์ด๋„๋ก ์„ค๊ณ„๋˜์—ˆ์Šต๋‹ˆ๋‹ค.
03:12
so that its motion can be recorded and interpreted as audio.
52
192928
4807
์ด ์›€์ง์ž„์€ ๊ธฐ๋กํ•  ์ˆ˜๋„ ์žˆ๊ณ  ์†Œ๋ฆฌ๋กœ ๋ณ€ํ™˜๋˜์–ด ์ฝํžˆ๊ธฐ๋„ ํ•ฉ๋‹ˆ๋‹ค.
03:17
But sound causes all objects to vibrate.
53
197735
3668
์†Œ๋ฆฌ๋Š” ์‚ฌ๋ฌผ์„ ์ง„๋™์‹œํ‚ฌ ์ˆ˜ ์žˆ์ง€๋งŒ,
03:21
Those vibrations are just usually too subtle and too fast for us to see.
54
201403
5480
์šฐ๋ฆฌ ๋ˆˆ์œผ๋กœ๋Š” ์ด ์ง„๋™์ด ๋งค์šฐ ๋ฏธ๋ฌ˜ํ•˜๊ณ  ๋นจ๋ผ์„œ ํ™•์ธํ•˜๊ธฐ ์–ด๋ ต์ฃ .
03:26
So what if we record them with a high-speed camera
55
206883
3738
๊ทธ๋ ‡๋‹ค๋ฉด ์šฐ๋ฆฌ๊ฐ€ ๊ณ ์† ์นด๋ฉ”๋ผ๋กœ ์˜์ƒ์„ ๊ธฐ๋กํ•˜๊ณ 
03:30
and then use software to extract tiny motions
56
210621
3576
์ด ์†Œํ”„ํŠธ์›จ์–ด๋ฅผ ์ด์šฉํ•ด ์นด๋ฉ”๋ผ๋กœ ์ฐ์€ ์•„์ฃผ ์ž‘์€ ๋ฏธ๋™์„ ์ถ”์ถœํ•ด ๋‚ธ๋’ค
03:34
from our high-speed video,
57
214197
2090
์–ด๋–ค ์†Œ๋ฆฌ๊ฐ€ ๊ทธ ์ง„๋™์„ ๋งŒ๋“ค์—ˆ๋Š”์ง€ ๋ถ„์„ํ•˜๋ฉด ์–ด๋–จ๊นŒ์š”?
03:36
and analyze those motions to figure out what sounds created them?
58
216287
4274
03:41
This would let us turn visible objects into visual microphones from a distance.
59
221859
5449
๋จผ๊ฑฐ๋ฆฌ์˜ ์‚ฌ๋ฌผ์„ ๋ณด๋Š” ๊ฒƒ ๋งŒ์œผ๋กœ๋„ ์†Œ๋ฆฌ๋ฅผ ์•Œ์•„๋‚ผ ์ˆ˜ ์žˆ๋„๋ก ํ•ด์ฃผ์ง€ ์•Š์„๊นŒ.
03:49
And so we tried this out,
60
229080
2183
๊ทธ๋ž˜์„œ ์ €ํฌ๊ฐ€ ์‹œ๋„ํ•ด ๋ณด์•˜์Šต๋‹ˆ๋‹ค.
03:51
and here's one of our experiments,
61
231263
1927
์ด๊ฒƒ์ด ์ €ํฌ๊ฐ€ ํ•œ ์‹คํ—˜ ์ค‘ ํ•˜๋‚˜์ธ๋ฐ์š”,
03:53
where we took this potted plant that you see on the right
62
233190
2949
ํ™”๋ฉด ์˜ค๋ฅธ์ชฝ์— ํ™”๋ถ„์„ ๊ฐ–๋‹ค ๋†“๊ณ 
03:56
and we filmed it with a high-speed camera
63
236139
2438
๊ทผ์ฒ˜์˜ ์Šคํ”ผ์ปค์— ์Œ์•…์„ ํฌ๊ฒŒ ํ‹€์–ด๋…ผ ๋’ค
03:58
while a nearby loudspeaker played this sound.
64
238577
3529
๊ณ ์†์นด๋ฉ”๋ผ๋กœ ์ดฌ์˜ํ•ด ๋ณด์•˜์Šต๋‹ˆ๋‹ค.
04:02
(Music: "Mary Had a Little Lamb")
65
242275
8190
(์Œ์•… : ๋–ด๋‹ค๋–ด๋‹ค ๋น„ํ–‰๊ธฐ) -์Šคํ”ผ์ปค๋ฅผ ํ†ตํ•ด ์Œ์•…์ด ๋‚˜์˜ด-
04:11
And so here's the video that we recorded,
66
251820
2824
์ด๊ฒƒ์ด ์ €ํฌ๊ฐ€ ์ดฌ์˜ํ•œ ์˜์ƒ์ž…๋‹ˆ๋‹ค.
04:14
and we recorded it at thousands of frames per second,
67
254644
3924
์ด ์˜์ƒ์€ ์ดˆ๋‹น ์ˆ˜์ฒœ ํ”„๋ ˆ์ž„์˜ ์†๋„๋กœ ๊ธฐ๋ก๋˜์—ˆ์ง€๋งŒ,
04:18
but even if you look very closely,
68
258568
2322
์—ฌ๋Ÿฌ๋ถ„์ด ์•„์ฃผ ๊ฐ€๊นŒ์ด์„œ ๋ณธ๋‹ค ํ•ด๋„
04:20
all you'll see are some leaves
69
260890
1951
๊ทธ๋ƒฅ ๊ฐ€๋งŒํžˆ ์žˆ๋Š”
04:22
that are pretty much just sitting there doing nothing,
70
262841
3065
๋‚˜๋ญ‡์žŽ๋“ค๋งŒ ๋ณด์ด์‹ค ๊ฒ๋‹ˆ๋‹ค.
04:25
because our sound only moved those leaves by about a micrometer.
71
265906
4806
์™œ๋ƒํ•˜๋ฉด ์ด ๋‚˜๋ญ‡์žŽ๋“ค์˜ ์›€์ง์ธ ๊ฑฐ๋ฆฌ๋Š” ๋งˆ์ดํฌ๋กœ๋ฏธํ„ฐ ์ •๋„๋กœ
04:31
That's one ten-thousandth of a centimeter,
72
271103
4276
1 ์„ผํ‹ฐ๋ฏธํ„ฐ์˜ ์ฒœ๋ถ„์˜ ์ผ ์ž…๋‹ˆ๋‹ค.
04:35
which spans somewhere between a hundredth and a thousandth
73
275379
4156
ํ™”๋ฉด์˜ 1 ํ™”์†Œ๋ฅผ ๋ฐฑ๋ถ„์˜ ์ผ์—์„œ ์ฒœ๋ถ„์˜ ์ผ๋กœ ๋‚˜๋ˆˆ ์ •๋„์ž…๋‹ˆ๋‹ค.
04:39
of a pixel in this image.
74
279535
2299
04:41
So you can squint all you want,
75
281881
2887
๊ทธ๋Ÿฌ๋‹ˆ ์—ฌ๋Ÿฌ๋ถ„ ๋งˆ์Œ๊ป ์งธ๋ ค๋ณด์„ธ์š”.
04:44
but motion that small is pretty much perceptually invisible.
76
284768
3335
๊ทธ๋Ÿฐ๋‹คํ•ด๋„ ์ด๋ ‡๊ฒŒ ๋ฏธ์„ธํ•œ ์›€์ง์ž„์€ ๋ˆˆ์œผ๋กœ ๋ถ„๋ณ„ํ•˜๊ธฐ ์–ด๋ ต์Šต๋‹ˆ๋‹ค.
04:49
But it turns out that something can be perceptually invisible
77
289667
4157
ํ•˜์ง€๋งŒ ์ด ์ž‘์€ ์›€์ง์ž„์€ ์œก์•ˆ์œผ๋กœ๋Š” ์ž๊ฐํ•˜๊ธฐ ์–ด๋ ค์šด ๊ฒƒ์ด์ง€๋งŒ
04:53
and still be numerically significant,
78
293824
2809
์ˆซ์ž์ ์œผ๋กœ๋Š” ๊ทธ ์˜๋ฏธ๊ฐ€ ์ถฉ๋ถ„ํžˆ ์žˆ๋Š” ๊ฒƒ์œผ๋กœ ๋“œ๋Ÿฌ๋‚ฌ์–ด์š”.
04:56
because with the right algorithms,
79
296633
2002
์™œ๋ƒํ•˜๋ฉด ์ œ๋Œ€๋กœ ๋œ ์•Œ๊ณ ๋ฆฌ๋“ฌ์„ ํ†ตํ•ด
04:58
we can take this silent, seemingly still video
80
298635
3687
๋ฌด์Œ์˜ ์ •์ง€ํ•ด ์žˆ๋Š” ๋“ฏํ•œ ๋™์˜์ƒ์„ ์ฐ์€ ๋’ค
๊ทธ ์˜์ƒ์œผ๋กœ๋ถ€ํ„ฐ ๋“ค์œผ์‹œ๋Š” ์†Œ๋ฆฌ๋ฅผ ๋ณต์›ํ•ด ๋‚ผ ์ˆ˜ ์žˆ์—ˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
05:02
and we can recover this sound.
81
302322
1527
05:04
(Music: "Mary Had a Little Lamb")
82
304690
7384
(์Œ์•…: ๋–ด๋‹ค๋–ด๋‹ค ๋น„ํ–‰๊ธฐ)
05:12
(Applause)
83
312074
5828
(๋ฐ•์ˆ˜)
05:22
So how is this possible?
84
322058
1939
์ด๊ฒƒ์ด ์–ด๋–ป๊ฒŒ ๊ฐ€๋Šฅํ•˜๋ƒ๊ณ ์š”?
05:23
How can we get so much information out of so little motion?
85
323997
4344
์ด ๋ฏธ์„ธํ•œ ์›€์ง์ž„์„ ํ†ตํ•ด ์ด๋ ‡๊ฒŒ๋‚˜ ๋ฐฉ๋Œ€ํ•œ ์ •๋ณด๋ฅผ ์–ป์„ ์ˆ˜ ์žˆ๋ƒ๊ณ ์š”?
05:28
Well, let's say that those leaves move by just a single micrometer,
86
328341
5361
์ด ๋‚˜๋ญ‡์žŽ์ด 1 ๋งˆ์ดํฌ๋กœ๋ฏธํ„ฐ๋งŒํผ ์›€์ง์ธ๋‹ค๊ณ  ํ•ฉ์‹œ๋‹ค.
05:33
and let's say that that shifts our image by just a thousandth of a pixel.
87
333702
4308
๊ทธ๋ฆฌ๊ณ  ๊ทธ๊ฒƒ์ด ์ฒœ๋งŒ๋ถ„์˜ 1 ํ™”์†Œ๋งŒํผ์˜ ์ด๋ฏธ์ง€๊ฐ€ ์ด๋™ํ–ˆ๋‹ค๊ณ  ํ•ฉ์‹œ๋‹ค.
05:39
That may not seem like much,
88
339269
2572
๊ทธ๋ฆฌ ํฐ ์ˆซ์ž ๊ฐ™์ง€๋Š” ์•Š์•„๋ณด์ž…๋‹ˆ๋‹ค.
05:41
but a single frame of video
89
341841
1996
ํ•˜์ง€๋งŒ ๋‹จ์ผ ํ”„๋ ˆ์ž„์˜ ๋น„๋””์˜ค๋Š”
05:43
may have hundreds of thousands of pixels in it,
90
343837
3257
๋ฐฑ๋งŒ๊ฐœ์— ๊ฐ€๊นŒ์šด ํ™”์†Œ๋กœ ์ด๋ฃจ์–ด์ ธ์žˆ๊ณ 
05:47
and so if we combine all of the tiny motions that we see
91
347094
3454
์ „์ฒด ์˜์ƒ์— ๊ฑธ์ณ ์ด๋ฅผ ๋ชจ๋‘ ํ•ฉ์น˜๋ฉด ์šฐ๋ฆฌ๊ฐ€ ๋ณผ ์ˆ˜ ์žˆ๋Š”
05:50
from across that entire image,
92
350548
2298
์•„์ฃผ ์ž‘์€ ์›€์ง์ž„์ด ๋ฉ๋‹ˆ๋‹ค.
05:52
then suddenly a thousandth of a pixel
93
352846
2623
๊ทธ๋Ÿฐํ›„์—” ์ด ์ฒœ๋ถ„์˜ 1 ํ™”์†Œ๊ฐ€
05:55
can start to add up to something pretty significant.
94
355469
2775
์ ์ธต์ ์œผ๋กœ ๋”ํ•ด์ ธ ์–ด๋–ค ์˜๋ฏธ์žˆ๋Š” ์›€์ง์ž„์œผ๋กœ ๋ฐ”๋€๋‹ˆ๋‹ค.
05:58
On a personal note, we were pretty psyched when we figured this out.
95
358870
3635
๊ฐœ์ธ์ ์œผ๋กœ๋Š”, ์šฐ๋ฆฌ๊ฐ€ ๋ฐํ˜€๋‚ธ ์ด ์‚ฌ์‹ค์— ๋Œ€ํ•ด ๋ชน์‹œ ํฅ๋ถ„ํ–ˆ์Šต๋‹ˆ๋‹ค.
06:02
(Laughter)
96
362505
2320
(์›ƒ์Œ)
06:04
But even with the right algorithm,
97
364825
3253
ํ•˜์ง€๋งŒ ์ œ๋Œ€๋กœ ๋œ ์•Œ๊ณ ๋ฆฌ๋“ฌ์„ ์ ์šฉํ–ˆ์–ด๋„
06:08
we were still missing a pretty important piece of the puzzle.
98
368078
3617
์•„์ง๊นŒ์ง€ ์ด ํผ์ฆ์˜ ๋งค์šฐ ์ค‘์š”ํ•œ ์กฐ๊ฐ์ด ์—†์—ˆ์Šต๋‹ˆ๋‹ค.
06:11
You see, there are a lot of factors that affect when and how well
99
371695
3604
๋ณด์‹œ๋‹ค ์‹œํ”ผ ๋งŽ์€ ์š”์†Œ๋“ค์ด ์–ธ์ œ, ์–ด๋–ป๊ฒŒ ์ด ๊ธฐ์ˆ ์ด
06:15
this technique will work.
100
375299
1997
์ž˜ ์ž‘๋™ํ•  ๊ฒƒ์ธ๊ฐ€์— ๋Œ€ํ•ด ์˜ํ–ฅ์„ ๋ผ์นฉ๋‹ˆ๋‹ค.
06:17
There's the object and how far away it is;
101
377296
3204
์ธก์ •ํ•˜๋ ค๋Š” ์‚ฌ๋ฌผ๊ณผ ๊ทธ ๊ฑฐ๋ฆฌ,
06:20
there's the camera and the lens that you use;
102
380500
2394
์–ด๋–ค ์นด๋ฉ”๋ผ์™€ ๋ Œ์ฆˆ๋ฅผ ์‚ฌ์šฉ ํ•  ์ง€,
06:22
how much light is shining on the object and how loud your sound is.
103
382894
4091
์–ผ๋งŒํผ์˜ ๋น›์„ ์‚ฌ๋ฌผ์— ๋…ธ์ถœํ•ด์•ผ ํ•  ์ง€ ์Œํ–ฅ์€ ์–ผ๋งˆ๋‚˜ ์ปค์•ผ ํ•˜๋Š”์ง€ ๋ง์ด์ฃ 
06:27
And even with the right algorithm,
104
387945
3375
๊ทธ๋ฆฌ๊ณ  ์ œ๋Œ€๋กœ ๋œ ์•Œ๊ณ ๋ฆฌ์ฆ˜์„ ์ด์šฉํ• ์ง€๋ผ๋„
06:31
we had to be very careful with our early experiments,
105
391320
3390
์ดˆ๊ธฐ์— ์‹คํ–‰๋œ ์‹คํ—˜์—์„œ๋Š” ๊นŠ์€ ์ฃผ์˜๋ฅผ ๊ธฐ์šธ์—ฌ์•ผ ํ–ˆ์Šต๋‹ˆ๋‹ค.
06:34
because if we got any of these factors wrong,
106
394710
2392
๋งŒ์•ฝ ์ด ์ค‘ ํ•˜๋‚˜๋ผ๋„ ์ž˜๋ชป๋œ ๊ฐ€์ •์ด ์žˆ์—ˆ๋‹ค๋ฉด
06:37
there was no way to tell what the problem was.
107
397102
2368
๋ฌด์—‡์ด ๋ฌธ์ œ์ธ์ง€ ์•Œ์•„๋‚ผ ๋ฐฉ๋ฒ•์ด ์—†์—ˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
06:39
We would just get noise back.
108
399470
2647
์•„๋งˆ๋„ ๊ทธ๋ƒฅ ์‹œ๋„๋Ÿฌ์šด ์†Œ์Œ๋งŒ ๊ฒฐ๊ณผ๋ฌผ๋กœ ์–ป์—ˆ๊ฒ ์ฃ .
06:42
And so a lot of our early experiments looked like this.
109
402117
3320
๊ทธ๋ž˜์„œ ๋งŽ์€ ์ดˆ๊ธฐ์˜ ์‹คํ—˜๋“ค์€ ๋‹ค์Œ๊ณผ ๊ฐ™์Šต๋‹ˆ๋‹ค.
06:45
And so here I am,
110
405437
2206
์—ฌ๊ธฐ ์ œ๊ฐ€ ์žˆ์Šต๋‹ˆ๋‹ค.
06:47
and on the bottom left, you can kind of see our high-speed camera,
111
407643
4040
ํ™”๋ฉด์•„๋ž˜ ์™ผ์ชฝ์— ์ดˆ๊ณ ์† ์นด๋ฉ”๋ผ๊ฐ€ ์–ธ๋œป ๋ณด์ด์‹œ์ฃ 
06:51
which is pointed at a bag of chips,
112
411683
2183
๊ฐ์ž์นฉ ๊ณผ์ž๋ด‰์ง€๋ฅผ ๋น„์ถ”๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
06:53
and the whole thing is lit by these bright lamps.
113
413866
2949
์ด ๋ชจ๋“  ๊ฒƒ์„ ๋น„์ถ”๋Š” ๊ฒƒ์ด ๋ฐ์€ ์ด ๋žจํ”„ ๋น›์ž…๋‹ˆ๋‹ค.
06:56
And like I said, we had to be very careful in these early experiments,
114
416815
4365
์ œ๊ฐ€ ๋ง์”€๋“œ๋ ธ๋“ฏ์ด ์ดˆ๊ธฐ ์‹คํ—˜์—์„œ๋Š” ๋ชจ๋“  ๊ฒƒ์— ๋Œ€ํ•ด ๋งค์šฐ ์กฐ์‹ฌ์Šค๋Ÿฌ์› ์Šต๋‹ˆ๋‹ค.
07:01
so this is how it went down.
115
421180
2508
์–ด๋–ป๊ฒŒ ์ง„ํ–‰๋˜์—ˆ๋Š”์ง€ ๋ณด์—ฌ๋“œ๋ฆด๊ฒŒ์š”.
07:03
(Video) Abe Davis: Three, two, one, go.
116
423688
3761
์…‹, ๋‘˜, ํ•˜๋‚˜, ์‹œ์ž‘
07:07
Mary had a little lamb! Little lamb! Little lamb!
117
427449
5387
"๋–ด๋‹ค ๋–ด๋‹ค ๋น„ํ–‰๊ธฐ! ๋‚ ์•„๋ผ, ๋‚ ์•„๋ผ! "
07:12
(Laughter)
118
432836
4500
(์›ƒ์Œ)
07:17
AD: So this experiment looks completely ridiculous.
119
437336
2814
๋งž์•„์š” ์ด ์‹คํ—˜์€ ์ •๋ง์ด์ง€ ์šฐ์Šค๊ฝ์Šค๋Ÿฌ์›Œ ๋ณด์ž…๋‹ˆ๋‹ค.
07:20
(Laughter)
120
440150
1788
(์›ƒ์Œ)
07:21
I mean, I'm screaming at a bag of chips --
121
441938
2345
๊ทธ๋Ÿฌ๋‹ˆ๊นŒ ์ €๋Š” ๊ณผ์ž๋ด‰์ง€์—๋‹ค ๋Œ€๊ณ  ์†Œ๋ฆฌ๋ฅผ ์ง€๋ฅด๊ณ 
07:24
(Laughter) --
122
444283
1551
(์›ƒ์Œ)
07:25
and we're blasting it with so much light,
123
445834
2117
์—„์ฒญ๋‚˜๊ฒŒ ๋ฐ์€ ์กฐ๋ช…์„ ์˜์•„๋Œ€์„œ
07:27
we literally melted the first bag we tried this on. (Laughter)
124
447951
4479
๋ง ๊ทธ๋Œ€๋กœ ์ฒซ๋ฒˆ์งธ ์‹คํ—˜ํ•œ ๊ณผ์ž๋ด‰์ง€๋ฅผ ๋…น์—ฌ๋ฒ„๋ฆด ์ •๋„์˜€์Šต๋‹ˆ๋‹ค. (์›ƒ์Œ)
07:32
But ridiculous as this experiment looks,
125
452525
3274
ํ•˜์ง€๋งŒ ์šฐ์Šค๊ฝ์Šค๋Ÿฝ๊ฒŒ ๋ณด์ด๋Š” ๋งŒํผ
07:35
it was actually really important,
126
455799
1788
๊ทธ๊ฒƒ์€ ๋งค์šฐ ์ค‘์š”ํ•œ ์‹คํ—˜์ด์—ˆ์–ด์š”.
07:37
because we were able to recover this sound.
127
457587
2926
์™œ๋ƒํ•˜๋ฉด ์ €ํฌ๋Š” ์Œํ–ฅ๋ณต์›์— ์„ฑ๊ณตํ–ˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
07:40
(Audio) Mary had a little lamb! Little lamb! Little lamb!
128
460513
4712
(์˜ค๋””์˜ค) ๋–ณ๋‹ค๋–ณ๋‹ค ๋น„ํ–‰๊ธฐ! ๋‚ ์•„๋ผ ๋‚ ์•„๋ผ!
07:45
(Applause)
129
465225
4088
(๋ฐ•์ˆ˜)
๊ทธ๋ฆฌ๊ณ  ์ด๋Š” ์ •๋ง์ด์ง€ ๋ง‰๋Œ€ํ•œ ์ค‘์š”์„ฑ์„ ๋•๋‹ˆ๋‹ค.
07:49
AD: And this was really significant,
130
469313
1881
07:51
because it was the first time we recovered intelligible human speech
131
471194
4119
์™œ๋ƒํ•˜๋ฉด ์ด ์‹คํ—˜์ด ์ตœ์ดˆ๋กœ ๋ฌด์Œ์˜ ๋™์˜์ƒ์—์„œ
07:55
from silent video of an object.
132
475424
2341
์ธ๊ฐ„์ด ๋งํ•˜๋Š” ์†Œ๋ฆฌ๋ฅผ ๋ณต์›ํ•ด ๋‚ธ ์‚ฌ๋ก€์ด๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
07:57
And so it gave us this point of reference,
133
477765
2391
์ด ์‹คํ—˜์„ ๊ธฐ๋ฐ˜์œผ๋กœ
08:00
and gradually we could start to modify the experiment,
134
480156
3871
์šฐ๋ฆฌ๋Š” ์ ์ฐจ ์‹คํ—˜์— ๋ณ€ํ˜•์„ ์‹œ๋„ํ–ˆ์Šต๋‹ˆ๋‹ค.
08:04
using different objects or moving the object further away,
135
484106
3805
๋‹ค์–‘ํ•œ ์‚ฌ๋ฌผ์„ ์ด์šฉํ•˜๊ฑฐ๋‚˜ ์ดฌ์˜ ๊ฑฐ๋ฆฌ๋ฅผ ๋” ๋ฉ€๋ฆฌ ์กฐ์ •ํ•˜๊ณ 
08:07
using less light or quieter sounds.
136
487911
2770
๋” ์ ์€ ์–‘์˜ ๋น›๊ณผ ๋” ์ž‘์€ ์†Œ๋ฆฌ๋ฅผ ์ด์šฉํ•˜๊ธฐ๋„ ํ–ˆ์Šต๋‹ˆ๋‹ค.
08:11
And we analyzed all of these experiments
137
491887
2874
์ด ๋‹ค์–‘ํ•œ ์‹คํ—˜๊ฒฐ๊ณผ๋“ค์„ ๋ถ„์„ํ•˜๋ฉฐ
08:14
until we really understood the limits of our technique,
138
494761
3622
์ด ๊ธฐ๋ฒ•์˜ ํ—ˆ์šฉ ํ•œ๋„๋ฅผ ์ดํ•ดํ•˜๊ฒŒ ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.
08:18
because once we understood those limits,
139
498383
1950
์™œ๋ƒํ•˜๋ฉด ์šฐ๋ฆฌ๊ฐ€ ์ด ํ•œ๋„๋ฅผ ์ดํ•ดํ•œ ๋’ค์—๋Š”
08:20
we could figure out how to push them.
140
500333
2346
๊ทธ ํ—ˆ์šฉ ํ•œ๋„๋ฅผ ์ดˆ์›”ํ•ด ๋ณผ ์ˆ˜ ์žˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
08:22
And that led to experiments like this one,
141
502679
3181
๊ทธ๋ž˜์„œ ๋‹ค์Œ๊ณผ ๊ฐ™์€ ์‹คํ—˜์„ ํ•˜๊ฒŒ ๋˜์—ˆ์Šต๋‹ˆ๋‹ค.
08:25
where again, I'm going to speak to a bag of chips,
142
505860
2739
๋˜ ๋‹ค์‹œ ์ €๋Š” ๊ณผ์ž๋ด‰์ง€์—๋‹ค ๋Œ€๊ณ  ์ด์•ผ๊ธฐ๋ฅผ ํ•ฉ๋‹ˆ๋‹ค.
08:28
but this time we've moved our camera about 15 feet away,
143
508599
4830
ํ•˜์ง€๋งŒ ์ด๋ฒˆ์—๋Š” ์นด๋ฉ”๋ผ๋ฅผ 4.5 ๋ฏธํ„ฐ ์ •๋„์˜ ๊ฑฐ๋ฆฌ๋กœ ์˜ฎ๊ธฐ๊ณ 
08:33
outside, behind a soundproof window,
144
513429
2833
๋ฐฉ์Œ์ด ๋˜๋Š” ์œ ๋ฆฌ์ฐฝ ๋’ค์— ์„ค์น˜ํ•˜์˜€์Šต๋‹ˆ๋‹ค.
08:36
and the whole thing is lit by only natural sunlight.
145
516262
2803
๋น›์ด๋ผ๊ณ ๋Š” ์ž์—ฐ๊ด‘์ด ์ „๋ถ€์ž…๋‹ˆ๋‹ค.
08:40
And so here's the video that we captured.
146
520529
2155
์ž ์ด๊ฒƒ์ด ์ €ํฌ๊ฐ€ ์ฐ์€ ๋™์˜์ƒ ์ž…๋‹ˆ๋‹ค.
08:44
And this is what things sounded like from inside, next to the bag of chips.
147
524450
4559
๋ฐฉ์Œ์ฐฝ ์•ˆ์— ์žˆ๋Š” ๊ณผ์ž๋ด‰์ง€ ์˜†์—์„œ ๋“ค๋ฆฌ๋Š” ์†Œ๋ฆฌ๋Š” ๋‹ค์Œ๊ณผ ๊ฐ™์Šต๋‹ˆ๋‹ค.
08:49
(Audio) Mary had a little lamb whose fleece was white as snow,
148
529009
5038
(์˜ค๋””์˜ค) ๋–ณ๋‹ค ๋–ณ๋‹ค ๋น„ํ–‰๊ธฐ ๋‚ ์•„๋ผ ๋‚ ์•„๋ผ
08:54
and everywhere that Mary went, that lamb was sure to go.
149
534047
5619
๋†’์ด๋†’์ด ๋‚ ์•„๋ผ ์šฐ๋ฆฌ๋น„ํ–‰๊ธฐ
08:59
AD: And here's what we were able to recover from our silent video
150
539666
4017
๊ทธ๋ฆฌ๊ณ  ์ด๊ฒƒ์ด ์šฐ๋ฆฌ๊ฐ€ ๋ฐฉ์Œ์ฐฝ ๋ฐ–์—์„œ ์ฐ์€
09:03
captured outside behind that window.
151
543683
2345
๋™์˜์ƒ์—์„œ ๋ณต์›ํ•ด ๋‚ธ ์Œํ–ฅ์ž…๋‹ˆ๋‹ค.
09:06
(Audio) Mary had a little lamb whose fleece was white as snow,
152
546028
4435
(์Œ์„ฑ) ๋–ด๋‹ค ๋–ด๋‹ค ๋น„ํ–‰๊ธฐ ๋‚ ์•„๋ผ ๋‚ ์•„๋ผ
09:10
and everywhere that Mary went, that lamb was sure to go.
153
550463
5457
๋†’์ด๋†’์ด ๋‚ ์•„๋ผ ์šฐ๋ฆฌ๋น„ํ–‰๊ธฐ
09:15
(Applause)
154
555920
6501
(๋ฐ•์ˆ˜)
09:22
AD: And there are other ways that we can push these limits as well.
155
562421
3542
์ด ํ—ˆ์šฉ ํ•œ๋„๋ฅผ ์ถ”์›”ํ•ด๋ณด๊ณ ์ž ์—ฌ๋Ÿฌ๊ฐ€์ง€ ๋‹ค๋ฅธ ๋ฐฉ๋ฒ•๋“ค๋„ ์‹œ๋„ํ–ˆ์Šต๋‹ˆ๋‹ค.
09:25
So here's a quieter experiment
156
565963
1798
๋‹ค์Œ์€ ์ข€ ๋” ์ž‘์€ ์†Œ๋ฆฌ๋ฅผ ์ด์šฉํ•œ ์‹คํ—˜์ž…๋‹ˆ๋‹ค.
09:27
where we filmed some earphones plugged into a laptop computer,
157
567761
4110
๋…ธํŠธ๋ถ ์ปดํ“จํ„ฐ์— ์—ฐ๊ฒฐํ•œ ์ด์–ดํฐ์„ ์ดฌ์˜ํ•œ ๊ฒƒ์œผ๋กœ
09:31
and in this case, our goal was to recover the music that was playing on that laptop
158
571871
4110
์ €ํฌ์˜ ๋ชฉํ‘œ๋Š” ์ด์–ดํฐ์—์„œ ํ˜๋Ÿฌ๋‚˜์˜ค๋Š” ์Œ์•…์„ ๋ณต์›ํ•ด ๋‚ด๋Š” ๊ฒƒ์ด์—ˆ์Šต๋‹ˆ๋‹ค.
09:35
from just silent video
159
575981
2299
๋ฌผ๋ก  ํ”Œ๋ผ์Šคํ‹ฑ ์ด์–ดํฐ์ด ์ฐํžŒ
09:38
of these two little plastic earphones,
160
578280
2507
์ด ๋™์˜์ƒ์€ ๋ฌด์Œ์ž…๋‹ˆ๋‹ค.
09:40
and we were able to do this so well
161
580787
2183
์ด ์‹คํ—˜์˜ ๊ฒฐ๊ณผ๋Š” ์ •ํ™•๋„๊ฐ€ ๋งค์šฐ ๋†’์•„
09:42
that I could even Shazam our results.
162
582970
2461
์ƒค์žผ(Shazam)์–ดํ”Œ์„ ํ†ตํ•ด ์Œ์•…์ฐพ๊ธฐ๋ฅผ ํ•  ์ˆ˜ ์žˆ์„ ์ •๋„์˜€์–ด์š”.
09:45
(Laughter)
163
585431
2411
(์›ƒ์Œ)
09:49
(Music: "Under Pressure" by Queen)
164
589191
10034
(์Œ์•…: "์–ธ๋” ํ”„๋ ˆ์…”" - ํ€ธ)
10:01
(Applause)
165
601615
4969
(๋ฐ•์ˆ˜)
10:06
And we can also push things by changing the hardware that we use.
166
606584
4551
๋‹ค์Œ์€ ๋‹ค๋ฅธ ์ข…๋ฅ˜์˜ ์žฅ๋น„๋ฅผ ์‚ฌ์šฉํ•ด์„œ ํ—ˆ์šฉ ํ•œ๋„๋ฅผ ์‹œํ—˜ํ•ด ๋ณด๊ธฐ๋„ ํ–ˆ์Šต๋‹ˆ๋‹ค.
10:11
Because the experiments I've shown you so far
167
611135
2461
์ง€๊ธˆ๊นŒ์ง€ ์ œ๊ฐ€ ๋ณด์—ฌ๋“œ๋ฆฐ ์‹คํ—˜๊ฒฐ๊ณผ๋“ค์€
10:13
were done with a camera, a high-speed camera,
168
613596
2322
์ดˆ๊ณ ์† ์นด๋ฉ”๋ผ๋ฅผ ์‚ฌ์šฉํ•œ ๊ฒƒ์ธ๋ฐ
10:15
that can record video about a 100 times faster
169
615918
2879
์ด ์นด๋ฉ”๋ผ๋Š” ์šฐ๋ฆฌ๊ฐ€ ๊ฐ€์ง„ ํ•ธ๋“œํฐ ์นด๋ฉ”๋ผ๋ณด๋‹ค
10:18
than most cell phones,
170
618797
1927
100 ๋ฐฐ๋‚˜ ๋น ๋ฅธ ๋…นํ™”๊ฐ€ ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค.
10:20
but we've also found a way to use this technique
171
620724
2809
ํ•˜์ง€๋งŒ ์ €ํฌ๋Š” ๋ณดํ†ต์˜ ์นด๋ฉ”๋ผ๋ฅผ ๊ฐ€์ง€๊ณ ๋„
10:23
with more regular cameras,
172
623533
2230
์ด๋ฅผ ๊ตฌํ˜„ํ•  ์ˆ˜ ์žˆ๋Š” ํ…Œํฌ๋‹‰์„ ์•Œ์•„๋ƒˆ์Šต๋‹ˆ๋‹ค.
10:25
and we do that by taking advantage of what's called a rolling shutter.
173
625763
4069
์ด๋ฅธ๋ฐ” "๋กค๋ง์…”ํ„ฐ"๋ผ ๋ถˆ๋ฆฌ์šฐ๋Š” ํšจ๊ณผ๋ฅผ ์ด์šฉํ•œ ๊ฒƒ์ธ๋ฐ์š”
10:29
You see, most cameras record images one row at a time,
174
629832
4798
๋งŽ์€ ์นด๋ฉ”๋ผ๋“ค์ด ์˜์ƒ์„ ํ•œ ๋ฒˆ์— ํ•œ ์ค„์”ฉ ๊ธฐ๋กํ•ฉ๋‹ˆ๋‹ค.
10:34
and so if an object moves during the recording of a single image,
175
634630
5702
๋งŒ์•ฝ ํ•œ ์žฅ๋ฉด ์ดฌ์˜์‹œ ์‚ฌ๋ฌผ์ด ์›€์ง์ด๋ฉด
10:40
there's a slight time delay between each row,
176
640344
2717
๊ฐ ์ค„ ์‚ฌ์ด ์‹œ๊ฐ„์ฐจ๊ฐ€ ์ƒ๊ธฐ๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
10:43
and this causes slight artifacts
177
643061
3157
์ด ๋•Œ๋ฌธ์— ์•ฝ๊ฐ„์˜ ์ธ์œ„์  ๋ณ€ํ˜•์ด ์ผ์–ด๋‚˜๊ฒŒ ๋˜๊ณ 
10:46
that get coded into each frame of a video.
178
646218
3483
์ด๊ฒƒ์ด ๋™์˜์ƒ ๊ฐ ํ”„๋ ˆ์ž„์— ๋‚จ์•„ ๊ธฐ๋ก๋ฉ๋‹ˆ๋‹ค.
10:49
And so what we found is that by analyzing these artifacts,
179
649701
3806
์šฐ๋ฆฌ๋Š” ์•Œ๊ณ ๋ฆฌ๋“ฌ์„ ๋ณ€ํ˜•ํ•˜์—ฌ ์ด ์ธ์œ„์  ๋ณ€ํ˜•์„ ๋ถ„์„ํ•œ ๊ฒฐ๊ณผ
10:53
we can actually recover sound using a modified version of our algorithm.
180
653507
4615
์ด ๋™์˜์ƒ์œผ๋กœ๋ถ€ํ„ฐ ์Œํ–ฅ์„ ๋ณต์›ํ•ด๋‚ผ ์ˆ˜ ์žˆ์—ˆ์Šต๋‹ˆ๋‹ค.
10:58
So here's an experiment we did
181
658122
1912
์ด๊ฒƒ์ด ์ €ํฌ๊ฐ€ ํ•œ ์‹คํ—˜์ž…๋‹ˆ๋‹ค.
11:00
where we filmed a bag of candy
182
660034
1695
๋ณด์‹œ๋Š” ๊ฒƒ์€ ์‚ฌํƒ•๋ด‰์ง€์ด๊ณ ์š”
11:01
while a nearby loudspeaker played
183
661729
1741
์ฃผ๋ณ€์— ์žˆ๋Š” ์Šคํ”ผ์ปค์—์„œ ํฐ ์†Œ๋ฆฌ๋กœ
11:03
the same "Mary Had a Little Lamb" music from before,
184
663470
2972
์ข…์ „๊ณผ ๊ฐ™์€ "๋–ณ๋‹ค ๋–ณ๋‹ค ๋น„ํ–‰๊ธฐ" ์Œ์•…์ด ํ˜๋Ÿฌ๋‚˜์˜ต๋‹ˆ๋‹ค.
11:06
but this time, we used just a regular store-bought camera,
185
666442
4203
ํ•˜์ง€๋งŒ ์ด๋ฒˆ์—๋Š” ์‹œ์ค‘์—์„œ ๊ตฌ์ž…ํ•œ ์ผ๋ฐ˜ ์นด๋ฉ”๋ผ๋ฅผ ์‚ฌ์šฉํ–ˆ์Šต๋‹ˆ๋‹ค.
11:10
and so in a second, I'll play for you the sound that we recovered,
186
670645
3174
์ž ์‹œ ํ›„ ์ €ํฌ๊ฐ€ ๋ณต์›ํ•œ ์†Œ๋ฆฌ๋ฅผ ๋“ค๋ ค๋“œ๋ฆดํ…๋ฐ์š”
11:13
and it's going to sound distorted this time,
187
673819
2050
์ด๋ฒˆ์—๋Š” ์•ฝ๊ฐ„ ๋’คํ‹€๋ฆฐ ๋“ฏํ•œ ์†Œ๋ฆฌ๋ฅผ ๋“ค์œผ์‹ค ๊ฒƒ์ž…๋‹ˆ๋‹ค.
11:15
but listen and see if you can still recognize the music.
188
675869
2836
ํ•˜์ง€๋งŒ ํ•œ ๋ฒˆ ๋“ค์–ด๋ณด์‹œ๊ณ  ๋ฌด์Šจ ์Œ์•…์ธ์ง€ ์•Œ ์ˆ˜ ์žˆ๋Š” ์ง€ ๋ณด์„ธ์š”.
11:19
(Audio: "Mary Had a Little Lamb")
189
679723
6223
(์˜ค๋””์˜ค: "๋–ณ๋‹ค๋–ณ๋‹ค ๋น„ํ–‰๊ธฐ")
11:37
And so, again, that sounds distorted,
190
697527
3465
์†Œ๋ฆฌ๋Š” ๋’คํ‹€๋ฆฐ ๋“ฏํ•˜์ง€๋งŒ ์ฃผ๋ชฉํ•˜์‹ค ์ ์€
11:40
but what's really amazing here is that we were able to do this
191
700992
4386
11:45
with something that you could literally run out
192
705378
2626
์†์‰ฝ๊ฒŒ ๊ตฌํ•  ์ˆ˜ ์žˆ๋Š” ๊ฒƒ๋“ค๋กœ
11:48
and pick up at a Best Buy.
193
708004
1444
์ด๋Ÿฌํ•œ ๊ธฐ์ˆ ์˜ ๊ตฌํ˜„์ด ๊ฐ€๋Šฅํ•˜๋‹ค๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
11:51
So at this point,
194
711122
1363
๊ทธ๋Ÿผ ์ด์ œ
11:52
a lot of people see this work,
195
712485
1974
๋งŽ์€ ์‚ฌ๋žŒ๋“ค์ด ์ด ์‹คํ—˜๊ฒฐ๊ณผ๋ฅผ ๋ณด๊ณ ๋Š”
11:54
and they immediately think about surveillance.
196
714459
3413
์ฆ‰๊ฐ์ ์œผ๋กœ "๊ฐ์‹œ์นด๋ฉ”๋ผ"๋ฅผ ๋– ์˜ฌ๋ฆฝ๋‹ˆ๋‹ค.
11:57
And to be fair,
197
717872
2415
๋„ค ๋งž์•„์š”. ๋ˆ„๊ตฐ๊ฐ€๋ฅผ ๊ฐ์‹œํ•˜๊ธฐ ์œ„ํ•ด
12:00
it's not hard to imagine how you might use this technology to spy on someone.
198
720287
4133
์ด ๊ธฐ์ˆ ์„ ์‚ฌ์šฉํ•œ ๋‹ค๋Š” ๊ฒƒ์„ ์ƒ์ƒํ•˜๊ธฐ๋ž€ ๊ทธ๋ฆฌ ์–ด๋ ต์ง€ ์•Š์Šต๋‹ˆ๋‹ค.
12:04
But keep in mind that there's already a lot of very mature technology
199
724420
3947
ํ•˜์ง€๋งŒ ํ˜„์žฌ์—๋„ ๊ฝค ์ˆ˜์ค€๋†’์€ ๊ฐ์‹œ์นด๋ฉ”๋ผ์™€ ์žฅ๋น„๋“ค์ด
12:08
out there for surveillance.
200
728367
1579
๋งŽ์ด ์žˆ๋‹ค๋Š” ๊ฒƒ์„ ์—ผ๋‘์— ๋‘์‹ญ์‹œ์š”.
12:09
In fact, people have been using lasers
201
729946
2090
์‚ฌ์‹ค, ๋งŽ์€ ์‚ฌ๋žŒ๋“ค์ด ๋ ˆ์ด์ €๋ฅผ ์ด์šฉํ•œ
12:12
to eavesdrop on objects from a distance for decades.
202
732036
2799
์›๊ฑฐ๋ฆฌ ๋„์ฒญ์„ ์ˆ˜์‹ญ๋…„๊ฐ„์ด๋‚˜ ํ•ด์™”์Šต๋‹ˆ๋‹ค.
12:15
But what's really new here,
203
735978
2025
ํ•˜์ง€๋งŒ ์—ฌ๊ธฐ์„œ ์ƒˆ๋กœ์šด ์ 
12:18
what's really different,
204
738003
1440
์ •๋ง๋กœ ๋‹ค๋ฅธ ์ ์€
12:19
is that now we have a way to picture the vibrations of an object,
205
739443
4295
์šฐ๋ฆฌ๋Š” ์ด์ œ ์‚ฌ๋ฌผ์˜ ์ง„๋™์„ ์‹œ๊ฐํ™” ํ•  ์ˆ˜ ์žˆ๋Š” ๋ฐฉ๋ฒ•์ด ์ƒ๊ฒผ๊ณ 
12:23
which gives us a new lens through which to look at the world,
206
743738
3413
๊ทธ ๊ธฐ์ˆ ์ด ์„ธ์ƒ์„ ๋‹ค๋ฅธ ๋ˆˆ์œผ๋กœ ๋ณผ ์ˆ˜ ์žˆ๊ฒŒ ํ•ด์ค€๋‹ค๋Š” ๊ฒ๋‹ˆ๋‹ค.
12:27
and we can use that lens
207
747151
1510
๋˜ํ•œ ์†Œ๋ฆฌ๋ฅผ ์ œ์–ดํ•˜์—ฌ ์ง„๋™์„ ์ผ์œผํ‚ค๋Š” ์š”์†Œ๊ฐ€ ๋ฌด์—‡์ธ์ง€ ๋ฟ ์•„๋‹ˆ๋ผ
12:28
to learn not just about forces like sound that cause an object to vibrate,
208
748661
4899
์‚ฌ๋ฌผ ๊ทธ ์ž์ฒด์˜ ์„ฑ์งˆ์— ๋Œ€ํ•ด์„œ๋„ ์•Œ ์ˆ˜ ์žˆ๊ฒŒ ํ•ด์ค๋‹ˆ๋‹ค.
12:33
but also about the object itself.
209
753560
2288
12:36
And so I want to take a step back
210
756975
1693
๊ทธ๋ž˜์„œ ์ €๋Š” ํ•œ๋ฐœ์ง ๋ฌผ๋Ÿฌ๋‚˜
12:38
and think about how that might change the ways that we use video,
211
758668
4249
์šฐ๋ฆฌ๊ฐ€ ๋™์˜์ƒ์„ ์ด์šฉํ•˜๋Š” ์šฉ๋„๊ฐ€
์–ด๋–ป๊ฒŒ ๋‹ฌ๋ผ์งˆ ์ˆ˜ ์žˆ๋Š”์ง€์— ๋Œ€ํ•ด ์ƒ๊ฐํ•ด ๋ณด๊ณ  ์‹ถ์–ด์š”.
12:42
because we usually use video to look at things,
212
762917
3553
์™œ๋ƒํ•˜๋ฉด ์šฐ๋ฆฌ๋Š” ์ฃผ๋กœ ์–ด๋–ค ๊ฒƒ๋“ค์„ ๋ณด๊ธฐ ์œ„ํ•ด ๋™์˜์ƒ์„ ์ด์šฉํ•˜๋Š”๋ฐ
12:46
and I've just shown you how we can use it
213
766470
2322
์ œ๊ฐ€ ๋ณด์—ฌ๋“œ๋ฆฐ ๊ฒƒ ์ฒ˜๋Ÿผ ์˜์ƒ์„ ํ†ตํ•ด
12:48
to listen to things.
214
768792
1857
๊ทธ๊ฒƒ์ด ๋‚ด๋Š” ์†Œ๋ฆฌ๋ฅผ ๋“ค์„ ์ˆ˜๋„ ์žˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
12:50
But there's another important way that we learn about the world:
215
770649
3971
ํ•˜์ง€๋งŒ ์‚ฌ๋ฌผ์˜ ์„ฑ์งˆ์— ๋Œ€ํ•ด ์•Œ ์ˆ˜ ์žˆ๋Š” ๋‹ค๋ฅธ ๋ฐฉ๋ฒ•์ด ์žˆ์Šต๋‹ˆ๋‹ค.
12:54
that's by interacting with it.
216
774620
2275
๋ฐ”๋กœ ์ง์ ‘ ์ž‘๋™ํ•ด ๋ณด๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
12:56
We push and pull and poke and prod things.
217
776895
3111
์šฐ๋ฆฌ๋Š” ์‚ฌ๋ฌผ์„ ๋ฐ€๊ธฐ๋„ ๋ถ™์žก๊ธฐ๊ณ  ํ•˜๊ณ  ์ฐŒ๋ฅด๊ฑฐ๋‚˜ ๋‹น๊ฒจ๋ณด๊ธฐ๋„ ํ•ฉ๋‹ˆ๋‹ค.
13:00
We shake things and see what happens.
218
780006
3181
ํ”๋“ค์–ด๋ณด๊ณ  ์–ด๋–ป๊ฒŒ ๋ฐ˜์‘ํ•˜๋Š”์ง€ ์‚ดํ”ผ๊ธฐ๋„ ํ•ฉ๋‹ˆ๋‹ค.
13:03
And that's something that video still won't let us do,
219
783187
4273
์ด๊ฒƒ์€ ์•„์ง๊นŒ์ง€๋„ ์šฐ๋ฆฌ๊ฐ€ ๋™์˜์ƒ์œผ๋กœ ํ•  ์ˆ˜ ์—†๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
13:07
at least not traditionally.
220
787460
2136
์ ์–ด๋„ ์ง€๊ธˆ๊นŒ์ง€ ์•Œ ๋˜ ๋ฐ”๋กœ๋Š” ๋ง์ด์ฃ .
13:09
So I want to show you some new work,
221
789596
1950
์ž ๊ทธ๋Ÿผ ์ƒˆ๋กœ์šด ํ”„๋กœ์ ํŠธ๋ฅผ ๋ณด์—ฌ๋“œ๋ฆฌ๊ฒ ์Šต๋‹ˆ๋‹ค.
13:11
and this is based on an idea I had just a few months ago,
222
791546
2667
์ด๋Š” ๋ช‡๋‹ฌ์ „์— ๋‚˜์˜จ ์•„์ด๋””์–ด๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ ํ•œ ๊ฒƒ์ธ๋ฐ,
13:14
so this is actually the first time I've shown it to a public audience.
223
794213
3301
์‹ค์ œ๋กœ ์˜ค๋Š˜ ์ฒ˜์Œ์œผ๋กœ ๋Œ€์ค‘์—๊ฒŒ ๊ณต๊ฐœํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
13:17
And the basic idea is that we're going to use the vibrations in a video
224
797514
5363
๋™์˜์ƒ์˜ ๋ฏธ๋™์„ ์ด์šฉํ•œ ๊ธฐ๋ณธ์ด๋ก ์„ ์ „์ œ๋กœ
13:22
to capture objects in a way that will let us interact with them
225
802877
4481
์‚ฌ๋ฌผ์ด ์šฐ๋ฆฌ์™€ ์ƒํ˜ธ ์ž‘์šฉ ํ•˜๋Š” ๋ฐฉ์‹์„ ํฌ์ฐฉํ•œ ๊ฒƒ์ธ๋ฐ์š”,
13:27
and see how they react to us.
226
807358
1974
์ด๋“ค์ด ์šฐ๋ฆฌ์—๊ฒŒ ์–ด๋–ป๊ฒŒ ๋ฐ˜์‘ ํ•˜๋Š”์ง€ ์•Œ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
13:31
So here's an object,
227
811120
1764
์ด๊ฒƒ์ด ์‚ฌ๋ฌผ์ž…๋‹ˆ๋‹ค.
13:32
and in this case, it's a wire figure in the shape of a human,
228
812884
3832
์ด ์‹คํ—˜์˜ ๊ฒฝ์šฐ ์‚ฌ๋žŒ๋ชจ์–‘์˜ ์ฒ ์‚ฌ๋กœ ๋งŒ๋“  ์ธํ˜•์ž…๋‹ˆ๋‹ค.
13:36
and we're going to film that object with just a regular camera.
229
816716
3088
์ผ๋ฐ˜ ์นด๋ฉ”๋ผ๋กœ ์ด ์‚ฌ๋ฌผ์„ ์ดฌ์˜ํ•ฉ๋‹ˆ๋‹ค.
13:39
So there's nothing special about this camera.
230
819804
2124
์นด๋ฉ”๋ผ ์ž์ฒด๋Š” ๋ณ„๋กœ ํŠน์ดํ•  ๊ฒƒ์ด ์—†์Šต๋‹ˆ๋‹ค.
13:41
In fact, I've actually done this with my cell phone before.
231
821928
2961
์‚ฌ์‹ค ์ด์ „์— ์ œ ํ•ธ๋“œํฐ ์นด๋ฉ”๋ผ๋กœ ์‹คํ—˜ํ•˜๊ธฐ๋„ ํ–ˆ์Šต๋‹ˆ๋‹ค.
13:44
But we do want to see the object vibrate,
232
824889
2252
์šฐ๋ฆฌ๋Š” ์‚ฌ๋ฌผ์˜ ์ง„๋™์„ ๊ด€์ฐฐํ•ด ๋ณด๊ณ ์ž ํ•˜๋Š”๋ฐ
13:47
so to make that happen,
233
827141
1133
๊ทธ๋Ÿฌ๊ธฐ ์œ„ํ•ด์„œ
13:48
we're just going to bang a little bit on the surface where it's resting
234
828274
3346
์‚ฌ๋ฌผ์ด ๋†“์—ฌ์ง„ ํ‘œ๋ฉด์„ ์„ธ๊ฒŒ ๋‘๋“ค๊ฒจ ๋ด…๋‹ˆ๋‹ค.
13:51
while we record this video.
235
831620
2138
์ดฌ์˜ํ•˜๋Š” ๋™์•ˆ ๋ง์ž…๋‹ˆ๋‹ค.
13:59
So that's it: just five seconds of regular video,
236
839398
3671
๊ทธ๋ ‡์Šต๋‹ˆ๋‹ค. ์šฐ๋ฆฌ๊ฐ€ ํ‘œ๋ฉด์„ ๋‘๋“ค๊ธฐ๋Š” ๋™์•ˆ ์ฐ์€
5์ดˆ ๊ธธ์ด์˜ ์ผ๋ฐ˜์ ์ธ ๋™์˜์ƒ ์ž…๋‹ˆ๋‹ค.
14:03
while we bang on this surface,
237
843069
2136
14:05
and we're going to use the vibrations in that video
238
845205
3513
์ง„๋™์ด ํฌ์ฐฉ๋œ ์ด ์˜์ƒ์„ ์ด์šฉํ•ด
14:08
to learn about the structural and material properties of our object,
239
848718
4544
์ด ์‚ฌ๋ฌผ์˜ ๊ตฌ์กฐ์™€ ๋ฌผ์งˆ์  ํŠน์ง•์ด ์–ด๋–ค ๊ฒƒ์ธ์ง€
14:13
and we're going to use that information to create something new and interactive.
240
853262
4834
๊ทธ ์ •๋ณด๋ฅผ ์ด์šฉํ•˜์—ฌ ์ง์ ‘ ์‚ฌ๋ฌผ์„ ์กฐ์ž‘ํ•ด ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
14:24
And so here's what we've created.
241
864866
2653
์ž ์ด๊ฒƒ์ด ์šฐ๋ฆฌ๊ฐ€ ๋งŒ๋“  ๊ฒƒ์ž…๋‹ˆ๋‹ค.
14:27
And it looks like a regular image,
242
867519
2229
๋ณด์‹œ๊ธฐ์—๋Š” ํ‰๋ฒ”ํ•œ ์‚ฌ์ง„ ๊ฐ™์Šต๋‹ˆ๋‹ค.
14:29
but this isn't an image, and it's not a video,
243
869748
3111
ํ•˜์ง€๋งŒ ์ด๊ฒƒ์€ ์‚ฌ์ง„๋„ ๋™์˜์ƒ๋„ ์•„๋‹™๋‹ˆ๋‹ค.
14:32
because now I can take my mouse
244
872859
2368
์™œ๋ƒํ•˜๋ฉด ์ง€๊ธˆ ์ œ๊ฐ€ ๋งˆ์šฐ์Šค๋ฅผ ๊ฐ–๋‹ค๋Œ€์„œ
14:35
and I can start interacting with the object.
245
875227
2859
์ด ์‚ฌ๋ฌผ์„ ์›€์ง์—ฌ ๋ณผ ์ˆ˜ ์žˆ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
14:44
And so what you see here
246
884936
2357
๋ณด์‹œ๋Š” ๊ฒƒ์€
์ด์ „์—๋Š” ๋ณด์ง€ ๋ชปํ–ˆ๋˜ ํž˜์„ ๊ฐ€ํ• ๋•Œ ์ด ์‚ฌ๋ฌผ์ด
14:47
is a simulation of how this object
247
887389
2226
14:49
would respond to new forces that we've never seen before,
248
889615
4458
์–ด๋–ป๊ฒŒ ๋ฐ˜์‘ํ•˜๋Š” ์ง€๋ฅผ ๋ณด์—ฌ์ฃผ๋Š” ์‹œ๋ฎฌ๋ ˆ์ด์…˜์ž…๋‹ˆ๋‹ค.
14:54
and we created it from just five seconds of regular video.
249
894073
3633
์˜ค์ง 5์ดˆ์งœ๋ฆฌ ์ผ๋ฐ˜ ๋™์˜์ƒ์„ ๊ฐ€์ง€๊ณ  ๋งŒ๋“ค์–ด๋‚ธ ๊ฒƒ ์ž…๋‹ˆ๋‹ค.
14:59
(Applause)
250
899249
4715
(๋ฐ•์ˆ˜)
15:09
And so this is a really powerful way to look at the world,
251
909421
3227
์ด๊ฒƒ์€ ์—„์ฒญ๋‚œ ์˜ํ–ฅ๋ ฅ์„ ์ง€๋‹Œ ์„ธ์ƒ์„ ๋ณด๋Š” ๋ฐฉ๋ฒ•์ž…๋‹ˆ๋‹ค.
15:12
because it lets us predict how objects will respond
252
912648
2972
์™œ๋ƒํ•˜๋ฉด ์ด๋กœ์จ ์‚ฌ๋ฌผ์ด ์ƒˆ๋กœ์šด ์ƒํ™ฉ์— ๋Œ€ํ•ด
์–ด๋–ป๊ฒŒ ๋ฐ˜์‘ํ• ์ง€ ์˜ˆ์ธกํ•  ์ˆ˜ ์žˆ๊ฒŒ ํ•ด์ฃผ๊ธฐ ๋•Œ๋ฌธ์ž…๋‹ˆ๋‹ค.
15:15
to new situations,
253
915620
1823
15:17
and you could imagine, for instance, looking at an old bridge
254
917443
3473
์˜ˆ๋ฅผ๋“ค์–ด ๋ณด์ฃ . ์—ฌ๋Ÿฌ๋ถ„์ด ๋‚ก์€ ๋‹ค๋ฆฌ๋ฅผ ๋ณด๊ณ 
15:20
and wondering what would happen, how would that bridge hold up
255
920916
3527
์ž๋™์ฐจ๋กœ ๊ทธ ๋‹ค๋ฆฌ๋ฅผ ๊ฑด๋„ ๋•Œ์—
15:24
if I were to drive my car across it.
256
924443
2833
๊ทธ ๋‹ค๋ฆฌ๊ฐ€ ์ž˜ ๋ฒ„ํ‹ธ์ง€๋ฅผ ๊ถ๊ธˆํ•ด ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
15:27
And that's a question that you probably want to answer
257
927276
2774
์ด๋Ÿฌํ•œ ์งˆ๋ฌธ์€ ๋ˆ„๊ตฌ๋ผ๋„ ๊ทธ ๋‹ต์„ ์•Œ๊ณ  ์‹ถ์–ด ํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
15:30
before you start driving across that bridge.
258
930050
2560
์‹ค์ œ๋กœ ์šด์ „ํ•ด์„œ ๋‹ค๋ฆฌ๋ฅผ ๊ฑด๋„ˆ๊ธฐ ์ „์— ๋ง์ž…๋‹ˆ๋‹ค.
15:33
And of course, there are going to be limitations to this technique,
259
933988
3272
๋ฌผ๋ก  ์•ž์„œ ์†Œ๊ฐœํ•ด๋“œ๋ฆฐ ์Œ์›๋ณต์› ๊ธฐ์ˆ ์ฒ˜๋Ÿผ
15:37
just like there were with the visual microphone,
260
937260
2462
์ด ๊ธฐ์ˆ ์—๋„ ํ•œ๊ณ„์ ์ด ์žˆ๊ฒ ์ง€๋งŒ
15:39
but we found that it works in a lot of situations
261
939722
3181
์šฐ๋ฆฌ๊ฐ€ ์˜ˆ์ƒ์น˜ ๋ชปํ•œ ๋งŽ์€ ์ƒํ™ฉ์—์„œ๋„
15:42
that you might not expect,
262
942903
1875
์ด ๊ธฐ์ˆ ์ด ์ž‘๋™ํ•œ๋‹ค๋Š” ๊ฒƒ์„ ํ™•์ธํ•˜์˜€์Šต๋‹ˆ๋‹ค.
15:44
especially if you give it longer videos.
263
944778
2768
ํŠนํžˆ ๋” ๊ธด ๊ธธ์ด์˜ ๋™์˜์ƒ์„ ์ด์šฉํ•˜๋ฉด ๋ง์ž…๋‹ˆ๋‹ค.
15:47
So for example, here's a video that I captured
264
947546
2508
๋ณด์‹œ๋Š” ๋™์˜์ƒ์€ ์ œ ์•„ํŒŒํŠธ ์•ž
15:50
of a bush outside of my apartment,
265
950054
2299
ํ™”๋‹จ์„ ์ดฌ์˜ํ•œ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
15:52
and I didn't do anything to this bush,
266
952353
3088
์ด ๋‚˜๋ญ‡๊ฐ€์ง€์— ๊ทธ ์–ด๋–ค ๊ฒƒ๋„ ํ•˜์ง€ ์•Š์•˜์Šต๋‹ˆ๋‹ค.
15:55
but by capturing a minute-long video,
267
955441
2705
1 ๋ถ„์ •๋„ ๊ธธ์ด์˜ ์ดฌ์˜๋งŒ์œผ๋กœ๋„
15:58
a gentle breeze caused enough vibrations
268
958146
3378
์•ฝํ•œ ๋ฐ”๋žŒ์— ์˜ํ•œ ์ง„๋™์ด ํฌ์ฐฉ๋˜์—ˆ๊ณ 
16:01
that we could learn enough about this bush to create this simulation.
269
961524
3587
์ด ํ™”๋‹จ์— ๋Œ€ํ•ด ์ถฉ๋ถ„ํ•œ ์ •๋ณด๋ฅผ ์–ป์–ด ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ๋งŒ๋“ค์–ด ๋ณผ ์ˆ˜ ์žˆ์—ˆ์Šต๋‹ˆ๋‹ค.
16:07
(Applause)
270
967270
6142
(๋ฐ•์ˆ˜)
16:13
And so you could imagine giving this to a film director,
271
973412
2972
์˜ํ™”๊ฐ๋…์—๊ฒŒ ์ด ์ด๋ฏธ์ง€๋ฅผ ์ค€๋‹ค๊ณ  ์ƒ์ƒํ•ด ๋ณด์„ธ์š”.
16:16
and letting him control, say,
272
976384
1719
์ด ์žฅ๋ฉด์ด ์ฐํžˆ๊ณ  ๋‚œ ํ›„์—
16:18
the strength and direction of wind in a shot after it's been recorded.
273
978103
4922
๋ฐ”๋žŒ์˜ ๊ฐ•๋„์™€ ๋ฐฉํ–ฅ์„ ์กฐ์ •ํ•  ์ˆ˜ ์žˆ๊ฒ ์ฃ .
16:24
Or, in this case, we pointed our camera at a hanging curtain,
274
984810
4535
๋ณด์‹œ๋Š” ๊ฒƒ์€ ๊ฑธ๋ ค ์žˆ๋Š” ์ปคํŠผ์„ ์ดฌ์˜ํ•œ ๊ฒƒ ์ž…๋‹ˆ๋‹ค.
16:29
and you can't even see any motion in this video,
275
989345
4129
๋™์˜์ƒ์—๋Š” ๋ˆˆ์œผ๋กœ ๋ณผ ์ˆ˜ ์žˆ๋Š” ํฐ ์›€์ง์ž„์ด ์—†์Šต๋‹ˆ๋‹ค.
16:33
but by recording a two-minute-long video,
276
993474
2925
ํ•˜์ง€๋งŒ 2๋ถ„์งœ๋ฆฌ ์˜์ƒ์„ ์ดฌ์˜ํ•จ์œผ๋กœ์จ
16:36
natural air currents in this room
277
996399
2438
๋ฐฉ์•ˆ์˜ ์ž์—ฐํ’์ด ๋งŒ๋“ค์–ด๋‚ด๋Š”
16:38
created enough subtle, imperceptible motions and vibrations
278
998837
4412
์•„์ฃผ ๋ฏธ์„ธํ•œ ์›€์ง์ž„๊ณผ ์ง„๋™์„ ํ†ตํ•ด
16:43
that we could learn enough to create this simulation.
279
1003249
2565
๋‹ค์Œ๊ณผ ๊ฐ™์€ ์‹œ๋ฎฌ๋ ˆ์ด์…˜์„ ๋งŒ๋“ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
16:48
And ironically,
280
1008243
2366
์•„์ด๋Ÿฌ๋‹ˆํ•˜๊ฒŒ๋„
16:50
we're kind of used to having this kind of interactivity
281
1010609
3088
์šฐ๋ฆฌ๋Š” ์ด๋ฏธ ๊ฐ€์ƒํ˜„์‹ค ๊ทธ๋ž˜ํ”ฝ์„ ํ†ตํ•ด ์ด๋Ÿฐ์‹์œผ๋กœ
16:53
when it comes to virtual objects,
282
1013697
2647
์ง์ ‘ ์กฐ์ž‘ํ•ด ๋ณด๋Š” ๊ฒƒ์— ๋Œ€ํ•ด ์ต์ˆ™ํ•ด์ ธ์žˆ์Šต๋‹ˆ๋‹ค.
16:56
when it comes to video games and 3D models,
283
1016344
3297
๋น„๋””์˜ค ๊ฒŒ์ž„์ด๋‚˜ 3D ๋ชจ๋ธ ๊ฐ™์€๊ฒƒ์ด์š”.
16:59
but to be able to capture this information from real objects in the real world
284
1019641
4404
ํ•˜์ง€๋งŒ ํ˜„์‹ค์„ธ๊ณ„์˜ ์‹ค์ œ ์‚ฌ๋ฌผ์„
17:04
using just simple, regular video,
285
1024045
2817
๋‹จ์ˆœํ•œ ๋™์˜์ƒ์„ ํ†ตํ•ด ์ด๋Ÿฌํ•œ ์ •๋ณด๋ฅผ ์–ป์–ด๋‚ด๋Š ๊ฒƒ์€
17:06
is something new that has a lot of potential.
286
1026862
2183
์ด์ „์—๋Š” ์—†์—ˆ๋˜ ๊ฒƒ์œผ๋กœ ๋งค์šฐ ํฐ ์ž ์žฌ๋ ฅ์„ ์ง€๋‹ˆ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
17:10
So here are the amazing people who worked with me on these projects.
287
1030410
4904
์ž, ์—ฌ๊ธฐ ์ด๋ถ„๋“ค์€ ์ด ํ”„๋กœ์ ํŠธ์— ์• ์จ์ฃผ์‹  ํ›Œ๋ฅญํ•œ ๋ถ„๋“ค์ž…๋‹ˆ๋‹ค.
17:16
(Applause)
288
1036057
5596
(๋ฐ•์ˆ˜)
17:24
And what I've shown you today is only the beginning.
289
1044819
3057
์ œ๊ฐ€ ์˜ค๋Š˜ ๋ณด์—ฌ๋“œ๋ฆฐ ๊ฒƒ์€ ๋‹จ์ง€ ์‹œ์ž‘์— ๋ถˆ๊ณผํ•ฉ๋‹ˆ๋‹ค.
17:27
We've just started to scratch the surface
290
1047876
2113
์ด๋Ÿฌํ•œ ์˜์ƒ์œผ๋กœ ํ•  ์ˆ˜ ์žˆ๋Š” ์ผ ์ค‘
17:29
of what you can do with this kind of imaging,
291
1049989
2972
๊ทนํžˆ ์ผ๋ถ€๋ถ„์— ๊ทผ์ ‘ํ–ˆ์„ ๋ฟ์ž…๋‹ˆ๋‹ค.
17:32
because it gives us a new way
292
1052961
2286
์ด ๊ธฐ์ˆ ์„ ํ†ตํ•ด ์šฐ๋ฆฌ ์ฃผ๋ณ€์˜ ๊ฒƒ๋“ค์„
17:35
to capture our surroundings with common, accessible technology.
293
1055342
4724
์ƒˆ๋กญ๊ฒŒ ๋ชจ์ƒ‰ํ•  ์ˆ˜ ์žˆ๋Š” ์ข€ ๋” ๋ณดํŽธ์ ์ธ ๋ฐฉ๋ฒ•์„ ๊ฐœ๋ฐœํ•  ๊ฒƒ์ž…๋‹ˆ๋‹ค.
17:40
And so looking to the future,
294
1060066
1929
๋ฏธ๋ž˜์—๋Š”
17:41
it's going to be really exciting to explore
295
1061995
2037
์ด ๊ธฐ์ˆ ์ด ๊ฐ€๋Šฅ์ผ€ ํ• 
17:44
what this can tell us about the world.
296
1064032
1856
์‹ ๋‚˜๋Š” ๋ชจํ—˜์ด ๊ธฐ๋‹ค๋ฆฌ๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
17:46
Thank you.
297
1066381
1204
๊ฐ์‚ฌํ•ฉ๋‹ˆ๋‹ค.
17:47
(Applause)
298
1067610
6107
(๋ฐ•์ˆ˜)
์ด ์›น์‚ฌ์ดํŠธ ์ •๋ณด

์ด ์‚ฌ์ดํŠธ๋Š” ์˜์–ด ํ•™์Šต์— ์œ ์šฉํ•œ YouTube ๋™์˜์ƒ์„ ์†Œ๊ฐœํ•ฉ๋‹ˆ๋‹ค. ์ „ ์„ธ๊ณ„ ์ตœ๊ณ ์˜ ์„ ์ƒ๋‹˜๋“ค์ด ๊ฐ€๋ฅด์น˜๋Š” ์˜์–ด ์ˆ˜์—…์„ ๋ณด๊ฒŒ ๋  ๊ฒƒ์ž…๋‹ˆ๋‹ค. ๊ฐ ๋™์˜์ƒ ํŽ˜์ด์ง€์— ํ‘œ์‹œ๋˜๋Š” ์˜์–ด ์ž๋ง‰์„ ๋”๋ธ” ํด๋ฆญํ•˜๋ฉด ๊ทธ๊ณณ์—์„œ ๋™์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค. ๋น„๋””์˜ค ์žฌ์ƒ์— ๋งž์ถฐ ์ž๋ง‰์ด ์Šคํฌ๋กค๋ฉ๋‹ˆ๋‹ค. ์˜๊ฒฌ์ด๋‚˜ ์š”์ฒญ์ด ์žˆ๋Š” ๊ฒฝ์šฐ ์ด ๋ฌธ์˜ ์–‘์‹์„ ์‚ฌ์šฉํ•˜์—ฌ ๋ฌธ์˜ํ•˜์‹ญ์‹œ์˜ค.

https://forms.gle/WvT1wiN1qDtmnspy7