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

204,360 views ใƒป 2015-05-05

TED


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

ืžืชืจื’ื: Ido Dekkers ืžื‘ืงืจ: Tal Dekkers
00:13
Most of us think of motion as a very visual thing.
0
13373
3349
ืจื•ื‘ื ื• ื—ื•ืฉื‘ื™ื ืขืœ ืชื ื•ืขื” ื›ืžืฉื”ื• ืžืื•ื“ ื•ื™ื–ื•ืืœื™.
00:17
If I walk across this stage or gesture with my hands while I speak,
1
17889
5088
ืื ืื ื™ ืืœืš ื‘ืจื—ื‘ื™ ื”ื‘ืžื” ืื• ืื—ื•ื•ื” ื‘ื™ื“ื™ื™ ื‘ืขื•ื“ื™ ืžื“ื‘ืจ,
00:22
that motion is something that you can see.
2
22977
2261
ื”ืชื ื•ืขื” ื”ื™ื ืžืฉื”ื• ืฉืืชื ื™ื›ื•ืœื™ื ืœืจืื•ืช.
00:26
But there's a world of important motion that's too subtle for the human eye,
3
26255
5482
ืื‘ืœ ื™ืฉ ืขื•ืœื ืฉืœ ืชื ื•ืขื•ืช ื—ืฉื•ื‘ื•ืช ืฉื”ืŸ ืขื“ื™ื ื•ืช ืžื“ื™ ืœืขื™ืŸ ื”ืื ื•ืฉื™ืช,
00:31
and over the past few years,
4
31737
2041
ื•ื‘ืžื”ืœืš ื”ืฉื ื™ื ื”ืื—ืจื•ื ื•ืช,
00:33
we've started to find that cameras
5
33778
1997
ื”ืชื—ืœื ื• ืœื’ืœื•ืช ืฉืžืฆืœืžื•ืช
00:35
can often see this motion even when humans can't.
6
35775
3410
ื™ื›ื•ืœื•ืช ื”ืจื‘ื” ืคืขืžื™ื ืœืจืื•ืช ืืช ื”ืชื ื•ืขื” ื”ื–ื• ืืคื™ืœื• ื›ืฉืื ืฉื™ื ืœื.
00:40
So let me show you what I mean.
7
40305
1551
ืื– ืชื ื• ืœื™ ืœื”ืจืื•ืช ืœื›ื ืœืžื” ืื ื™ ืžืชื›ื•ื•ืŸ.
00:42
On the left here, you see video of a person's wrist,
8
42717
3622
ืžืฆื“ ืฉืžืืœ ืคื”, ืืชื ืจื•ืื™ื ืกืจื˜ื•ืŸ ืฉืœ ืคืจืง ื›ืฃ ื™ื“ ืฉืœ ืื“ื,
00:46
and on the right, you see video of a sleeping infant,
9
46339
3147
ื•ืžื™ืžื™ืŸ ืืชื ืจื•ืื™ื ืกืจื˜ื•ืŸ ืฉืœ ืชื™ื ื•ืงืช ื™ืฉื ื”,
00:49
but if I didn't tell you that these were videos,
10
49486
3146
ืื‘ืœ ืื ืœื ื”ื™ื™ืชื™ ืื•ืžืจ ืœื›ื ืฉืืœื” ืกืจื˜ื•ื ื™ื,
00:52
you might assume that you were looking at two regular images,
11
52632
3761
ืื•ืœื™ ื”ื™ื™ืชื ืžื ื™ื—ื™ื ืฉืืชื ืžื‘ื™ื˜ื™ื ื‘ืฉืชื™ ืชืžื•ื ื•ืช ืจื’ื™ืœื•ืช,
00:56
because in both cases,
12
56393
1672
ืžืคื ื™ ืฉื‘ืฉื ื™ ื”ืžืงืจื™ื,
00:58
these videos appear to be almost completely still.
13
58065
3047
ื”ืกืจื˜ื•ื ื™ื ื”ืืœื” ื ืจืื™ื ื›ืžืขื˜ ืœื’ืžืจื™ ื ื™ื™ื—ื™ื.
01:02
But there's actually a lot of subtle motion going on here,
14
62175
3885
ืื‘ืœ ื™ืฉ ืœืžืขืฉื” ื”ืจื‘ื” ืชื ื•ืขื•ืช ืขื“ื™ื ื•ืช ืฉืžืชืจื—ืฉื•ืช ืฉื,
01:06
and if you were to touch the wrist on the left,
15
66060
2392
ื•ืื ื”ื™ื™ืชื ื ื•ื’ืขื™ื ื‘ืคืจืง ื”ื™ื“ ืžืฉืžืืœ,
01:08
you would feel a pulse,
16
68452
1996
ื”ื™ื™ืชื ืžืจื’ื™ืฉื™ื ื“ื•ืคืง,
01:10
and if you were to hold the infant on the right,
17
70448
2485
ื•ืื ื”ื™ื™ืชื ืžื—ื–ื™ืงื™ื ืืช ื”ืชื™ื ื•ืงืช ืžื™ืžื™ืŸ,
01:12
you would feel the rise and fall of her chest
18
72933
2391
ื”ื™ื™ืชื ืžืจื’ื™ืฉื™ื ืืช ื”ืขืœื™ื” ื•ื”ื™ืจื™ื“ื” ืฉืœ ื”ื—ื–ื” ืฉืœื”
01:15
as she took each breath.
19
75324
1390
ื›ืฉื”ื™ื ืœื•ืงื—ืช ื›ืœ ื ืฉื™ืžื”.
01:17
And these motions carry a lot of significance,
20
77762
3576
ื•ื”ืชื ื•ืขื•ืช ื”ืืœื• ื‘ืขืœื•ืช ืžืฉืžืขื•ืช ื’ื“ื•ืœื” ืžืื•ื“,
01:21
but they're usually too subtle for us to see,
21
81338
3343
ืื‘ืœ ื”ืŸ ื‘ื“ืจืš ื›ืœืœ ืขื“ื™ื ื•ืช ืžื“ื™ ื›ื“ื™ ืฉื ื•ื›ืœ ืœืจืื•ืช ืื•ืชืŸ,
01:24
so instead, we have to observe them
22
84681
2276
ืื– ื‘ืžืงื•ื, ืื ื—ื ื• ืฆืจื™ื›ื™ื ืœืฆืคื•ืช ื‘ื”ืŸ
01:26
through direct contact, through touch.
23
86957
2900
ื“ืจืš ืžื’ืข ื™ืฉื™ืจ, ื“ืจืš ืžื’ืข.
01:30
But a few years ago,
24
90997
1265
ืื‘ืœ ืœืคื ื™ ื›ืžื” ืฉื ื™ื,
01:32
my colleagues at MIT developed what they call a motion microscope,
25
92262
4405
ื”ืขืžื™ืชื™ื ืฉืœื™ ื‘ MIT ืคื™ืชื—ื• ืžื” ืฉื”ื ืงื•ืจืื™ื ืœื• ืžื™ืงืจื•ืกืงื•ืค ืชื ื•ืขื”,
01:36
which is software that finds these subtle motions in video
26
96667
4384
ืฉื”ื•ื ืชื•ื›ื ื” ืฉืžื’ืœื” ืืช ื”ืชื ื•ืขื•ืช ื”ืขื“ื™ื ื•ืช ื”ืืœื• ื‘ืกืจื˜ื•ืŸ
01:41
and amplifies them so that they become large enough for us to see.
27
101051
3562
ื•ืžื’ื‘ื™ืจื” ืื•ืชืŸ ื›ืš ืฉื”ืŸ ื”ื•ืคื›ื•ืช ืœื’ื“ื•ืœื•ืช ืžืกืคื™ืง ื›ื“ื™ ืฉื ืจืื” ืื•ืชืŸ.
01:45
And so, if we use their software on the left video,
28
105416
3483
ื•ื›ืš, ืื ื ืฉืชืžืฉ ื‘ืชื•ื›ื ื” ืฉืœื”ื ืขืœ ื”ืกืจื˜ื•ืŸ ื”ืฉืžืืœื™,
01:48
it lets us see the pulse in this wrist,
29
108899
3250
ื–ื” ื ื•ืชืŸ ืœื ื• ืœืจืื•ืช ืืช ื”ื“ื•ืคืง ื‘ืคืจืง ื”ื™ื“,
01:52
and if we were to count that pulse,
30
112149
1695
ื•ืื ื”ื™ื™ื ื• ืกื•ืคืจื™ื ืืช ื”ื“ื•ืคืง ื”ื–ื”,
01:53
we could even figure out this person's heart rate.
31
113844
2355
ื ื•ื›ืœ ืืคื™ืœื• ืœื“ืขืช ืžื” ื”ื“ื•ืคืง ืฉืœ ื”ืื“ื ื”ื–ื”.
01:57
And if we used the same software on the right video,
32
117095
3065
ื•ืื ื”ื™ื™ื ื• ืžืฉืชืฉืžื™ื ื‘ืื•ืชื” ืชื•ื›ื ื” ืขืœ ื”ืกืจื˜ื•ืŸ ื”ื™ืžื ื™,
02:00
it lets us see each breath that this infant takes,
33
120160
3227
ื–ื” ื ื•ืชืŸ ืœื ื• ืœืจืื•ืช ื›ืœ ื ืฉื™ืžื” ืฉื”ืชื™ื ื•ืงืช ื”ื–ื• ืœื•ืงื—ืช,
02:03
and we can use this as a contact-free way to monitor her breathing.
34
123387
4137
ื•ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ื–ื” ื›ื“ืจืš ื ื˜ื•ืœืช ืžื’ืข ื›ื“ื™ ืœื ื˜ืจ ืืช ื”ื ืฉื™ืžื” ืฉืœื”.
02:08
And so this technology is really powerful because it takes these phenomena
35
128884
5348
ื•ื›ืš ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ื”ื–ื• ื”ื™ื ืžืžืฉ ื—ื–ืงื” ืžืคื ื™ ืฉื”ื™ื ืœื•ืงื—ืช ืืช ื”ืชื•ืคืขื” ื”ื–ื•
02:14
that we normally have to experience through touch
36
134232
2367
ืฉืื ื—ื ื• ื‘ื“ืจืš ื›ืœืœ ื—ื•ื•ื™ื ื“ืจืš ืžื’ืข
02:16
and it lets us capture them visually and non-invasively.
37
136599
2957
ื•ื”ื™ื ืžืืคืฉืจืช ืœื ื• ืœืœื›ื•ื“ ืื•ืชื” ื•ื™ื–ื•ืืœื™ืช ื•ืœืœื ืคืœื™ืฉื”.
02:21
So a couple years ago, I started working with the folks that created that software,
38
141104
4411
ืื– ืœืคื ื™ ื›ืžื” ืฉื ื™ื, ื”ืชื—ืœืชื™ ืœืขื‘ื•ื“ ืขื ื”ืื ืฉื™ื ืฉื™ืฆืจื• ืืช ื”ืชื•ื›ื ื” ื”ื–ื•,
02:25
and we decided to pursue a crazy idea.
39
145515
3367
ื•ื”ื—ืœื˜ื ื• ืœื ืกื•ืช ืจืขื™ื•ืŸ ืžืฉื•ื’ืข.
02:28
We thought, it's cool that we can use software
40
148882
2693
ื—ืฉื‘ื ื•, ื–ื” ืžื’ื ื™ื‘ ืฉืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ืชื•ื›ื ื”
02:31
to visualize tiny motions like this,
41
151575
3135
ื›ื“ื™ ืœื“ืžื•ืช ืชื ื•ืขื•ืช ื–ืขื™ืจื•ืช ื›ืžื• ืืœื•,
02:34
and you can almost think of it as a way to extend our sense of touch.
42
154710
4458
ื•ืืชื ื™ื›ื•ืœื™ื ื›ืžืขื˜ ืœื—ืฉื•ื‘ ืขืœ ื–ื” ื›ื“ืจืš ืœื”ืจื—ื™ื‘ ืืช ื—ื•ืฉ ื”ืžื’ืข ืฉืœื ื•.
02:39
But what if we could do the same thing with our ability to hear?
43
159168
4059
ืื‘ืœ ืžื” ืื ื ื•ื›ืœ ืœืขืฉื•ืช ืื•ืชื• ื”ื“ื‘ืจ ืขื ื”ื™ื›ื•ืœืช ืœืฉืžื•ืข?
02:44
What if we could use video to capture the vibrations of sound,
44
164508
4665
ืžื” ืื ื ื•ื›ืœ ืœื”ืฉืชืžืฉ ื‘ืกืจื˜ื•ืŸ ื›ื“ื™ ืœืœื›ื•ื“ ืืช ื”ืจืขื™ื“ื•ืช ืฉืœ ื”ืฆืœื™ืœื™ื,
02:49
which are just another kind of motion,
45
169173
2827
ืฉื”ืŸ ืคืฉื•ื˜ ืขื•ื“ ืกื•ื’ ืฉืœ ืชื ื•ืขื”,
02:52
and turn everything that we see into a microphone?
46
172000
3346
ื•ืœื”ืคื•ืš ื›ืœ ืžื” ืฉืื ื—ื ื• ืจื•ืื™ื ืœืžื™ืงืจื•ืคื•ืŸ?
02:56
Now, this is a bit of a strange idea,
47
176236
1971
ืขื›ืฉื™ื•, ื–ื” ืจืขื™ื•ืŸ ืžืขื˜ ืžื•ื–ืจ,
02:58
so let me try to put it in perspective for you.
48
178207
2586
ืื– ืชื ื• ืœื™ ืœืฉื™ื ืืช ื–ื” ื‘ืคืจืกืคืงื˜ื™ื‘ื” ื‘ืฉื‘ื™ืœื›ื.
03:01
Traditional microphones work by converting the motion
49
181523
3488
ืžื™ืงืจื•ืคื•ื ื™ื ืžืกื•ืจืชื™ื™ื ืขื•ื‘ื“ื™ื ืขืœ ื™ื“ื™ ื”ืžืจืช ื”ืชื ื•ืขื”
03:05
of an internal diaphragm into an electrical signal,
50
185011
3599
ืฉืœ ื“ื™ืืคืจื’ืžื” ืคื ื™ืžื™ืช ืœืื•ืช ื—ืฉืžืœื™,
03:08
and that diaphragm is designed to move readily with sound
51
188610
4318
ื•ื”ื“ื™ืืคืจื’ืžื” ืžืชื•ื›ื ื ืช ืœื ื•ืข ื‘ืงืœื•ืช ืขื ืงื•ืœ
03:12
so that its motion can be recorded and interpreted as audio.
52
192928
4807
ื›ืš ืฉื”ืชื ื•ืขื” ืชื•ื›ืœ ืœื”ื™ื•ืช ืžื•ืงืœื˜ืช ื•ืžืชื•ืจื’ืžืช ืœืื•ื“ื™ื•.
03:17
But sound causes all objects to vibrate.
53
197735
3668
ืื‘ืœ ืฆืœื™ืœื™ื ื’ื•ืจืžื™ื ืœื›ืœ ื”ืขืฆืžื™ื ืœืจื˜ื•ื˜.
03:21
Those vibrations are just usually too subtle and too fast for us to see.
54
201403
5480
ื”ืจื˜ื™ื˜ื•ืช ื”ืืœื• ื”ืŸ ืคืฉื•ื˜ ื‘ื“ืจืš ื›ืœืœ ืขื“ื™ื ื•ืช ื•ืžื”ื™ืจื•ืช ืžื“ื™ ื›ื“ื™ ืฉื ื•ื›ืœ ืœืจืื•ืช ืื•ืชืŸ.
03:26
So what if we record them with a high-speed camera
55
206883
3738
ืื– ืžื” ืื ื ืงืœื™ื˜ ืื•ืชืŸ ืขื ืžืฆืœืžื” ื‘ืžื”ื™ืจื•ืช ื’ื‘ื•ื”ื”
03:30
and then use software to extract tiny motions
56
210621
3576
ื•ืื– ื ืฉืชืžืฉ ื‘ืชื•ื›ื ื” ื›ื“ื™ ืœืœื›ื•ื“ ืชื ื•ื“ื•ืช ื–ืขื™ืจื•ืช
03:34
from our high-speed video,
57
214197
2090
ืžื”ื•ื™ื“ืื• ื‘ืžื”ื™ืจื•ืช ื’ื‘ื•ื”ื” ืฉืœื ื•,
03:36
and analyze those motions to figure out what sounds created them?
58
216287
4274
ื•ื ื ืชื— ืืช ื”ืชื ื•ืขื•ืช ื”ืืœื• ื›ื“ื™ ืœื”ื‘ื™ืŸ ืื™ื–ื” ืฆืœื™ืœื™ื ื™ืฆืจื• ืื•ืชืŸ?
03:41
This would let us turn visible objects into visual microphones from a distance.
59
221859
5449
ื–ื” ื™ืชืŸ ืœื ื• ืœื”ืคื•ืš ืขืฆืžื™ื ื ืจืื™ื ืœืžื™ืงืจื•ืคื•ื ื™ื ื•ื™ื–ื•ืืœื™ื™ื ืžืžืจื—ืง.
03:49
And so we tried this out,
60
229080
2183
ื•ื›ืš ื ื™ืกื™ื ื• ืืช ื–ื”,
03:51
and here's one of our experiments,
61
231263
1927
ื•ื”ื ื” ืื—ื“ ื”ื ื™ืกื•ื™ื™ื,
03:53
where we took this potted plant that you see on the right
62
233190
2949
ื‘ื• ืœืงื—ื ื• ืฆืžื— ื‘ืขืฆื™ืฅ ืฉืืชื ืจื•ืื™ื ืžื™ืžื™ืŸ
03:56
and we filmed it with a high-speed camera
63
236139
2438
ื•ืฆื™ืœืžื ื• ืื•ืชื• ืขื ืžืฆืœืžื” ื‘ืžื”ื™ืจื•ืช ื’ื‘ื•ื”ื”
03:58
while a nearby loudspeaker played this sound.
64
238577
3529
ื‘ืขื•ื“ ืจืžืงื•ืœ ื‘ื™ืชื™ ื ื™ื’ืŸ ืืช ื”ืฆืœื™ืœ ื”ื–ื”.
04:02
(Music: "Mary Had a Little Lamb")
65
242275
8190
(ืžื•ื–ื™ืงื”: "ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ")
04:11
And so here's the video that we recorded,
66
251820
2824
ื•ื”ื ื” ื”ืกืจื˜ื•ืŸ ืฉื”ืงืœื˜ื ื•,
04:14
and we recorded it at thousands of frames per second,
67
254644
3924
ื•ื”ืงืœื˜ื ื• ืืช ื–ื” ื‘ืืœืคื™ ืคืจื™ื™ืžื™ื ื‘ืฉื ื™ื”,
04:18
but even if you look very closely,
68
258568
2322
ืื‘ืœ ืืคื™ืœื• ืื ืชื‘ื™ื˜ื• ืžืงืจื•ื‘,
04:20
all you'll see are some leaves
69
260890
1951
ื›ืœ ืžื” ืฉืืชื ืจื•ืื™ื ื–ื” ื›ืžื” ืขืœื™ื
04:22
that are pretty much just sitting there doing nothing,
70
262841
3065
ืฉืคืฉื•ื˜ ื™ื•ืฉื‘ื™ื ืฉื ื•ืœื ืขื•ืฉื™ื ื›ืœื•ื,
04:25
because our sound only moved those leaves by about a micrometer.
71
265906
4806
ืžืคื ื™ ืฉื”ืฆืœื™ืœ ืฉืœื ื• ื”ื–ื™ื– ืืช ื”ืขืœื™ื ื”ืืœื” ื‘ืขืจืš ื‘ืžื™ืงืจื• ืžื˜ืจ.
04:31
That's one ten-thousandth of a centimeter,
72
271103
4276
ื–ื” ืขืฉื™ืจื™ืช ืืœืคื™ืช ืฉืœ ืกื ื˜ื™ืžื˜ืจ,
04:35
which spans somewhere between a hundredth and a thousandth
73
275379
4156
ืฉืžืชืจื—ื‘ืช ืœืžืื™ืช ืื• ืืœืคื™ืช
04:39
of a pixel in this image.
74
279535
2299
ืฉืœ ืคื™ืงืกืœ ื‘ืชืžื•ื ื” ื”ื–ื•.
04:41
So you can squint all you want,
75
281881
2887
ืื– ืืชื ื™ื›ื•ืœื™ื ืœืžืฆืžืฅ ื›ืžื” ืฉืืชื ืจื•ืฆื™ื,
04:44
but motion that small is pretty much perceptually invisible.
76
284768
3335
ืื‘ืœ ืชื ื•ืขื” ื›ืœ ื›ืš ืงื˜ื ื” ื”ื™ื ื‘ืœืชื™ ื ืจืื™ืช ืชืคื™ืกืชื™ืช.
04:49
But it turns out that something can be perceptually invisible
77
289667
4157
ืื‘ืœ ืžืกืชื‘ืจ ืฉืžืฉื”ื• ื™ื›ื•ืœ ืœื”ื™ื•ืช ื‘ืœืชื™ ื ืจืื” ืชืคื™ืกืชื™ืช
04:53
and still be numerically significant,
78
293824
2809
ื•ืขื“ื™ื™ืŸ ืœื”ื™ื•ืช ืžืฉืžืขื•ืชื™ืช ืžืกืคืจื™ืช,
04:56
because with the right algorithms,
79
296633
2002
ืžืคื ื™ ืฉืขื ื”ืืœื’ื•ืจื™ืชืžื™ื ื”ื ื›ื•ื ื™ื,
04:58
we can take this silent, seemingly still video
80
298635
3687
ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืงื—ืช ืืช ื”ืกืจื˜ื•ืŸ ื”ืฉืงื˜ ื”ื–ื”, ืฉื ืจืื” ื“ื•ืžื
05:02
and we can recover this sound.
81
302322
1527
ื•ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืฉื—ื–ืจ ืืช ื”ืฆืœื™ืœ.
05:04
(Music: "Mary Had a Little Lamb")
82
304690
7384
(ืžื•ื–ื™ืงื”: "ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ")
05:12
(Applause)
83
312074
5828
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
05:22
So how is this possible?
84
322058
1939
ืื– ืื™ืš ื–ื” ืืคืฉืจื™?
05:23
How can we get so much information out of so little motion?
85
323997
4344
ืื™ืš ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืงื‘ืœ ื›ืœ ื›ืš ื”ืจื‘ื” ืžื™ื“ืข ืžื›ืœ ื›ืš ืžืขื˜ ืชื ื•ืขื”?
05:28
Well, let's say that those leaves move by just a single micrometer,
86
328341
5361
ื•ื‘ื›ืŸ, ื‘ื•ืื• ื ื’ื™ื“ ืฉื”ืขืœื™ื ื”ืืœื” ื–ื–ื™ื ื‘ืžื™ืงืจื•ืžื˜ืจ ืื—ื“ ื‘ืœื‘ื“,
05:33
and let's say that that shifts our image by just a thousandth of a pixel.
87
333702
4308
ื•ื‘ื•ืื• ื ื’ื™ื“ ืฉื–ื” ืžื–ื™ื– ืืช ื”ืชืžื•ื ื” ืฉืœื ื• ืจืง ื‘ืืœืคื™ืช ืคื™ืงืกืœ.
05:39
That may not seem like much,
88
339269
2572
ื–ื” ืื•ืœื™ ืœื ื ืจืื” ื”ืจื‘ื”,
05:41
but a single frame of video
89
341841
1996
ืื‘ืœ ื‘ืคืจื™ื™ื ื™ื—ื™ื“ ืฉืœ ืกืจื˜ื•ืŸ
05:43
may have hundreds of thousands of pixels in it,
90
343837
3257
ื™ืฉ ืื•ืœื™ ืžืื•ืช ืืœืคื™ ืคื™ืงืกืœื™ื,
05:47
and so if we combine all of the tiny motions that we see
91
347094
3454
ื•ื›ืš ืื ื ืฉืœื‘ ืืช ื›ืœ ื”ืชื ื•ืขื•ืช ื”ื–ืขื™ืจื•ืช ื”ืืœื• ืฉืื ื—ื ื• ืจื•ืื™ื
05:50
from across that entire image,
92
350548
2298
ื‘ืจื—ื‘ื™ ื›ืœ ื”ืชืžื•ื ื”,
05:52
then suddenly a thousandth of a pixel
93
352846
2623
ืื– ืคืชืื•ื ืืœืคื™ืช ืคื™ืงืกืœ
05:55
can start to add up to something pretty significant.
94
355469
2775
ื™ื›ื•ืœื” ืœื”ืชื•ื•ืกืฃ ืœืžืฉื”ื• ื“ื™ ืžืฉืžืขื•ืชื™.
05:58
On a personal note, we were pretty psyched when we figured this out.
95
358870
3635
ื‘ื ื™ืžื” ืื™ืฉื™ืช, ื“ื™ ื”ืชืœื”ื‘ื ื• ื›ืฉื”ื‘ื ื• ืืช ื–ื”.
06:02
(Laughter)
96
362505
2320
(ืฆื—ื•ืง)
06:04
But even with the right algorithm,
97
364825
3253
ืื‘ืœ ืืคื™ืœื• ืขื ื”ืืœื’ื•ืจื™ืชื ื”ื ื›ื•ืŸ,
06:08
we were still missing a pretty important piece of the puzzle.
98
368078
3617
ืขื“ื™ื™ืŸ ื”ื™ื” ื—ืกืจ ืœื ื• ื—ืœืง ื“ื™ ื—ืฉื•ื‘ ืฉืœ ื”ืคืื–ืœ.
06:11
You see, there are a lot of factors that affect when and how well
99
371695
3604
ืืชื ืžื‘ื™ื ื™ื, ื™ืฉ ื”ืจื‘ื” ืžืฉืชื ื™ื ืฉืžืฉืคื™ืขื™ื ืขืœ ืžืชื™ ื•ื›ืžื” ื˜ื•ื‘
06:15
this technique will work.
100
375299
1997
ื”ืฉื™ื˜ื” ื”ื–ื• ืชืขื‘ื•ื“.
06:17
There's the object and how far away it is;
101
377296
3204
ื™ืฉ ืืช ื”ืขืฆื ื•ื›ืžื” ืจื—ื•ืง ื”ื•ื;
06:20
there's the camera and the lens that you use;
102
380500
2394
ื™ืฉ ืืช ื”ืžืฆืœืžื” ื•ื”ืขื“ืฉื” ื‘ื”ื ืืชื ืžืฉืชืžืฉื™ื;
06:22
how much light is shining on the object and how loud your sound is.
103
382894
4091
ื›ืžื” ืื•ืจ ืžืื™ืจ ืขืœ ื”ืขืฆื ื•ื›ืžื” ื—ื–ืง ื”ืฆืœื™ืœ.
06:27
And even with the right algorithm,
104
387945
3375
ื•ืืคื™ืœื• ืขื ื”ืืœื’ื•ืจื™ืชื ื”ื ื›ื•ืŸ,
06:31
we had to be very careful with our early experiments,
105
391320
3390
ื”ื™ื™ื ื• ืฆืจื™ื›ื™ื ืœื”ื™ื•ืช ืžืื•ื“ ื–ื”ื™ืจื™ื ืขื ื”ื ื™ืกื•ื™ื™ื ื”ืจืืฉื•ื ื™ื,
06:34
because if we got any of these factors wrong,
106
394710
2392
ืžืคื ื™ ืฉืื ืื—ื“ ื”ืžืฉืชื ื™ื ื”ืืœื” ื”ื™ื” ืฉื’ื•ื™
06:37
there was no way to tell what the problem was.
107
397102
2368
ืœื ื”ื™ืชื” ื“ืจืš ืœื“ืขืช ืžื” ื”ื‘ืขื™ื”.
06:39
We would just get noise back.
108
399470
2647
ืคืฉื•ื˜ ื”ื™ื™ื ื• ืžืงื‘ืœื™ื ืจืขืฉ.
06:42
And so a lot of our early experiments looked like this.
109
402117
3320
ื•ื›ืš ื”ืจื‘ื” ืžื”ื ื™ืกื•ื™ื™ื ื”ืจืืฉื•ื ื™ื ื ืจืื• ื›ืš.
06:45
And so here I am,
110
405437
2206
ืื– ื”ื ื” ืื ื™,
06:47
and on the bottom left, you can kind of see our high-speed camera,
111
407643
4040
ื•ืžืฉืžืืœ ืœืžื˜ื”, ืืชื ืกื•ื’ ืฉืœ ื™ื›ื•ืœื™ื ืœืจืื•ืช ืืช ื”ืžืฆืœืžื” ื”ืžื”ื™ืจื” ืฉืœื ื•,
06:51
which is pointed at a bag of chips,
112
411683
2183
ืฉืžืฆื‘ื™ืขื” ืขืœ ืฉืงื™ืช ืฆ'ื™ืคืก,
06:53
and the whole thing is lit by these bright lamps.
113
413866
2949
ื•ื›ืœ ื–ื” ืžื•ืืจ ืขืœ ื™ื“ื™ ื”ืžื ื•ืจื•ืช ื”ื‘ื•ื”ืงื•ืช ื”ืืœื•.
06:56
And like I said, we had to be very careful in these early experiments,
114
416815
4365
ื•ื›ืžื• ืฉืืžืจืชื™, ื”ื™ื™ื ื• ืฆืจื™ื›ื™ื ืœื”ื™ื•ืช ืžืื•ื“ ื–ื”ื™ืจื™ื ื‘ื ื™ืกื•ื™ื™ื ื”ืจืืฉื•ื ื™ื ื”ืืœื”,
07:01
so this is how it went down.
115
421180
2508
ืื– ื›ืš ื–ื” ื”ืชืจื—ืฉ.
07:03
(Video) Abe Davis: Three, two, one, go.
116
423688
3761
(ืกืจื˜ื•ืŸ)ืื™ื™ืก ื“ื™ื™ื•ื™ืก: ืฉืœื•ืฉ, ืฉืชื™ื™ื, ืื—ื“, ื’ื•.
07:07
Mary had a little lamb! Little lamb! Little lamb!
117
427449
5387
ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ! ื˜ืœื” ืงื˜ืŸ! ื˜ืœื” ืงื˜ืŸ!
07:12
(Laughter)
118
432836
4500
(ืฆื—ื•ืง)
07:17
AD: So this experiment looks completely ridiculous.
119
437336
2814
ื.ื“: ืื– ื”ื ื™ืกื•ื™ ื”ื–ื” ื ืจืื” ืžื’ื•ื—ืš ืœื’ืžืจื™.
07:20
(Laughter)
120
440150
1788
(ืฆื—ื•ืง)
07:21
I mean, I'm screaming at a bag of chips --
121
441938
2345
ืื ื™ ืžืชื›ื•ื•ืŸ, ืื ื™ ืฆื•ืจื— ืขืœ ืฉืงื™ืช ืฆื™'ืคืก --
07:24
(Laughter) --
122
444283
1551
(ืฆื—ื•ืง) --
07:25
and we're blasting it with so much light,
123
445834
2117
ื•ืื ื—ื ื• ืžืคืฆื™ืฆื™ื ืื•ืชื” ื‘ื›ืœ ื›ืš ื”ืจื‘ื” ืื•ืจ,
07:27
we literally melted the first bag we tried this on. (Laughter)
124
447951
4479
ืฉืœืžืขืฉื” ื”ืžืกื ื• ืืช ื”ืฉืงื™ืช ื”ืจืืฉื•ื ื” ืฉื ื™ืกื™ื ื• ืขืœื™ื” ืืช ื–ื”. (ืฆื—ื•ืง)
07:32
But ridiculous as this experiment looks,
125
452525
3274
ืื‘ืœ ืžื’ื•ื—ืš ื›ื›ืœ ืฉื–ื” ื ืฉืžืข,
07:35
it was actually really important,
126
455799
1788
ื–ื” ื”ื™ื” ืœืžืขืฉื” ืžืื•ื“ ื—ืฉื•ื‘,
07:37
because we were able to recover this sound.
127
457587
2926
ืžืคื ื™ ืฉื”ื™ื™ื ื• ืžืกื•ื’ืœื™ื ืœืฉื—ื–ืจ ืืช ื”ืฆืœื™ืœ ื”ื–ื”.
07:40
(Audio) Mary had a little lamb! Little lamb! Little lamb!
128
460513
4712
(ืื•ื“ื™ื•) ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ! ื˜ืœื” ืงื˜ืŸ! ื˜ืœื” ืงื˜ืŸ!
07:45
(Applause)
129
465225
4088
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
07:49
AD: And this was really significant,
130
469313
1881
ื.ื“: ื•ื–ื” ื”ื™ื” ืžืžืฉ ืžืฉืžืขื•ืชื™,
07:51
because it was the first time we recovered intelligible human speech
131
471194
4119
ืžืคื ื™ ืฉื–ื• ื”ื™ืชื” ื”ืคืขื ื”ืจืืฉื•ื ื” ืฉืฉื—ื–ืจื ื• ื“ื™ื‘ื•ืจ ืื“ื ืžื•ื‘ืŸ
07:55
from silent video of an object.
132
475424
2341
ืžืกืจื˜ื•ืŸ ื“ื•ืžื ืฉืœ ืขืฆื.
07:57
And so it gave us this point of reference,
133
477765
2391
ืื– ื–ื” ื ืชืŸ ืœื ื• ื ืงื•ื“ืช ื”ืชื™ื™ื—ืกื•ืช,
08:00
and gradually we could start to modify the experiment,
134
480156
3871
ื•ืœื‘ืกื•ืฃ ื ื•ื›ืœ ืœื”ืชื—ื™ืœ ืœืฉื ื•ืช ืืช ื”ื ื™ืกื•ื™,
08:04
using different objects or moving the object further away,
135
484106
3805
ื‘ืฉื™ืžื•ืฉ ื‘ืขืฆืžื™ื ืฉื•ื ื™ื ืื• ืœื”ืจื—ื™ืง ืืช ื”ืขืฆืžื™ื,
08:07
using less light or quieter sounds.
136
487911
2770
ืฉื™ืžื•ืฉ ื‘ืคื—ื•ืช ืื•ืจ ืื• ืฆืœื™ืœื™ื ืฉืงื˜ื™ื ื™ื•ืชืจ.
08:11
And we analyzed all of these experiments
137
491887
2874
ื•ื ื™ืชื—ื ื• ืืช ื›ืœ ื”ื ื™ืกื•ื™ื™ื ื”ืืœื”
08:14
until we really understood the limits of our technique,
138
494761
3622
ืขื“ ืฉื‘ืืžืช ื”ื‘ื ื• ืืช ื”ืžื’ื‘ืœื•ืช ืฉืœ ื”ืฉื™ื˜ื” ืฉืœื ื•,
08:18
because once we understood those limits,
139
498383
1950
ืžืคื ื™ ืฉื‘ืจื’ืข ืฉื”ื‘ื ื• ืืช ื”ืžื’ื‘ืœื•ืช ื”ืืœื•,
08:20
we could figure out how to push them.
140
500333
2346
ื™ื›ื•ืœื ื• ืœื”ื‘ื™ืŸ ืื™ืš ืœื“ื—ื•ืฃ ืื•ืชืŸ.
08:22
And that led to experiments like this one,
141
502679
3181
ื•ื–ื” ื”ื•ื‘ื™ืœ ืœื ื™ืกื•ื™ื™ื ื›ืžื• ื–ื”,
08:25
where again, I'm going to speak to a bag of chips,
142
505860
2739
ื›ืฉืฉื•ื‘, ืื ื™ ืขื•ืžื“ ืœื“ื‘ืจ ืœืฉืงื™ืช ืฆ'ื™ืคืก,
08:28
but this time we've moved our camera about 15 feet away,
143
508599
4830
ืื‘ืœ ื”ืคืขื ื”ื–ื–ื ื• ืืช ื”ืžืฆืœืžื” ืœืžืจื—ืง 5 ืžื˜ืจ,
08:33
outside, behind a soundproof window,
144
513429
2833
ื‘ื—ื•ืฅ, ืžืื—ื•ืจื™ ื—ืœื•ืŸ ืื˜ื•ื ืœืฆืœื™ืœื™ื,
08:36
and the whole thing is lit by only natural sunlight.
145
516262
2803
ื•ื›ืœ ื–ื” ืžื•ืืจ ื‘ืื•ืจ ืฉืžืฉ ื˜ื‘ืขื™.
08:40
And so here's the video that we captured.
146
520529
2155
ื•ื›ืš ื”ื ื” ื”ืกืจื˜ื•ืŸ ืฉืœื›ื“ื ื•.
08:44
And this is what things sounded like from inside, next to the bag of chips.
147
524450
4559
ื•ื›ืš ื ืฉืžืขื• ื”ื“ื‘ืจื™ื ืžื‘ืคื ื™ื, ืœื™ื“ ืฉืงื™ืช ื”ืฆ'ื™ืคืก.
08:49
(Audio) Mary had a little lamb whose fleece was white as snow,
148
529009
5038
(ืื•ื“ื™ื•) ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ ืฉืฆืžืจื• ื”ื™ื” ืœื‘ืŸ ื›ืฉืœื’,
08:54
and everywhere that Mary went, that lamb was sure to go.
149
534047
5619
ื•ืœื›ืœ ืžืงื•ื ืฉืžืจื™ ื”ืœื›ื”, ื”ื˜ืœื” ื”ื™ื” ืื™ืชื”.
08:59
AD: And here's what we were able to recover from our silent video
150
539666
4017
ื.ื“: ื•ื”ื ื” ืžื” ืฉื”ื™ื™ื ื• ืžืกื•ื’ืœื™ื ืœืฉื—ื–ืจ ืžื”ืกืจื˜ื•ืŸ ื”ื“ื•ืžื ืฉืœื ื•
09:03
captured outside behind that window.
151
543683
2345
ืฉืฆื•ืœื ื‘ื—ื•ืฅ ืžืื—ื•ืจื™ ื”ื—ืœื•ืŸ ื”ื”ื•ื.
09:06
(Audio) Mary had a little lamb whose fleece was white as snow,
152
546028
4435
(ืื•ื“ื™ื•) ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ ืฉืฆืžืจื• ื”ื™ื” ืœื‘ืŸ ื›ืฉืœื’,
09:10
and everywhere that Mary went, that lamb was sure to go.
153
550463
5457
ื•ืœื›ืœ ืžืงื•ื ืฉืžืจื™ ื”ืœื›ื”, ื”ื˜ืœื” ื”ื™ื” ืื™ืชื”.
09:15
(Applause)
154
555920
6501
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
09:22
AD: And there are other ways that we can push these limits as well.
155
562421
3542
ื.ื“: ื•ื™ืฉ ื“ืจื›ื™ื ืื—ืจื•ืช ืฉื’ื ืื™ืชืŸ ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื“ื—ื•ืฃ ืืช ื”ืžื’ื‘ืœื•ืช.
09:25
So here's a quieter experiment
156
565963
1798
ืื– ื”ื ื” ื ื™ืกื•ื™ ืฉืงื˜ ื™ื•ืชืจ
09:27
where we filmed some earphones plugged into a laptop computer,
157
567761
4110
ื‘ื• ืฆื™ืœืžื ื• ืื•ื–ื ื™ื•ืช ืžื—ื•ื‘ืจื•ืช ืœืžื—ืฉื‘ ื ื™ื™ื“,
09:31
and in this case, our goal was to recover the music that was playing on that laptop
158
571871
4110
ื•ื‘ืžืงืจื” ื”ื–ื”, ื”ืžื˜ืจื” ืฉืœื ื• ื”ื™ืชื” ืœืฉื—ื–ืจ ืืช ื”ืžื•ื–ื™ืงื” ืฉื ื•ื’ื ื” ื‘ืžื—ืฉื‘ ื”ื ื™ื™ื“
09:35
from just silent video
159
575981
2299
ืžืกืจื˜ื•ืŸ ื“ื•ืžื ื‘ืœื‘ื“
09:38
of these two little plastic earphones,
160
578280
2507
ืฉืœ ืฉืชื™ ื”ืื•ื–ื ื™ื•ืช ื”ืคืœืกื˜ื™ื•ืช ื”ืงื˜ื ื•ืช ื”ืืœื•,
09:40
and we were able to do this so well
161
580787
2183
ื•ื”ื™ื™ื ื• ืžืกื•ื’ืœื™ื ืœืขืฉื•ืช ื–ืืช ื›ืœ ื›ืš ื˜ื•ื‘
09:42
that I could even Shazam our results.
162
582970
2461
ืฉืืคื™ืœื• ื™ื›ื•ืœืชื™ ืœื”ืจื™ืฅ ืืช ื”ืชื•ืฆืื” ื‘ืฉื–ืื.
09:45
(Laughter)
163
585431
2411
(ืฆื—ื•ืง)
09:49
(Music: "Under Pressure" by Queen)
164
589191
10034
(ืžื•ื–ื™ืงื”: "ืชื—ืช ืœื—ืฅ" ืฉืœ ืงื•ื•ื™ืŸ)
10:01
(Applause)
165
601615
4969
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
10:06
And we can also push things by changing the hardware that we use.
166
606584
4551
ื•ืื ื—ื ื• ื™ื›ื•ืœื™ื ื’ื ืœื“ื—ื•ืฃ ื“ื‘ืจื™ื ืขืœ ื™ื“ื™ ืฉื™ื ื•ื™ ื”ื—ื•ืžืจื” ื‘ื” ืื ื—ื ื• ืžืฉืชืžืฉื™ื.
10:11
Because the experiments I've shown you so far
167
611135
2461
ืžืคื ื™ ืฉื”ื ื™ืกื•ื™ื ืฉื”ืจืืชื™ ืขื“ ืขื›ืฉื™ื•
10:13
were done with a camera, a high-speed camera,
168
613596
2322
ื ืขืฉื• ืขื ืžืฆืœืžื”, ืžืฆืœืžื” ืžื”ื™ืจื”,
10:15
that can record video about a 100 times faster
169
615918
2879
ืฉื™ื›ื•ืœื” ืœื”ืงืœื™ื˜ ืกืจื˜ื•ื ื™ื ื‘ืžื”ื™ืจื•ืช ื’ื“ื•ืœื” ืคื™ 100 ื‘ืขืจืš
10:18
than most cell phones,
170
618797
1927
ืžืจื•ื‘ ื”ื˜ืœืคื•ื ื™ื ื”ืกืœื•ืœื•ืจื™ื™ื,
10:20
but we've also found a way to use this technique
171
620724
2809
ืื‘ืœ ื’ื™ืœื™ื ื• ื’ื ื“ืจืš ืœื”ืฉืชืžืฉ ื‘ื˜ื›ื ื™ืงื”
10:23
with more regular cameras,
172
623533
2230
ืขื ืžืฆืœืžื•ืช ืจื’ื™ืœื•ืช ื™ื•ืชืจ,
10:25
and we do that by taking advantage of what's called a rolling shutter.
173
625763
4069
ื•ืื ื—ื ื• ืขื•ืฉื™ื ืืช ื–ื” ืขืœ ื™ื“ื™ ื ื™ืฆื•ืœ ืžื” ืฉื ืงืจื ืฆืžืฆื ืžืชื’ืœื’ืœ.
10:29
You see, most cameras record images one row at a time,
174
629832
4798
ืืชื ืžื‘ื™ื ื™ื, ืจื•ื‘ ื”ืžืฆืœืžื•ืช ืžืงืœื™ื˜ื•ืช ืชืžื•ื ื•ืช ืฉื•ืจื” ืื—ืช ืื—ืจื™ ื”ืฉื ื™ื”,
10:34
and so if an object moves during the recording of a single image,
175
634630
5702
ื•ื›ืš ืื ื”ืขืฆื ื ืข ื‘ืžื”ืœืš ื”ืงืœื˜ืช ืชืžื•ื ื” ืื—ืช,
10:40
there's a slight time delay between each row,
176
640344
2717
ื™ืฉ ืขื™ื›ื•ื‘ ื–ืžืŸ ืงื˜ืŸ ื‘ื™ืŸ ื›ืœ ืฉื•ืจื”,
10:43
and this causes slight artifacts
177
643061
3157
ื•ื–ื” ื’ื•ืจื ืœืคื’ืžื™ื ืงืœื™ื
10:46
that get coded into each frame of a video.
178
646218
3483
ืฉืžืงื•ื“ื“ื™ื ืœื›ืœ ืคืจื™ื™ื ืฉืœ ื”ืกืจื˜ื•ืŸ.
10:49
And so what we found is that by analyzing these artifacts,
179
649701
3806
ื•ื›ืš ืžื” ืฉื’ื™ืœื™ื ื• ื–ื” ืฉืขืœ ื™ื“ื™ ื ื™ืชื•ื— ื”ืคื’ืžื™ื ื”ืืœื”,
10:53
we can actually recover sound using a modified version of our algorithm.
180
653507
4615
ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืžืขืฉื” ืœืฉื—ื–ืจ ืฆืœื™ืœ ื‘ืฉื™ืžื•ืฉ ื‘ื’ืจืกื” ืžืขื•ื“ื›ื ืช ืฉืœ ื”ืืœื’ื•ืจื™ืชื ืฉืœื ื•.
10:58
So here's an experiment we did
181
658122
1912
ืื– ื”ื ื” ื“ื•ื’ืžื” ืฉืขืฉื™ื ื•
11:00
where we filmed a bag of candy
182
660034
1695
ื‘ื” ืฆื™ืœืžื ื• ืฉืงื™ืช ืกื•ื›ืจื™ื•ืช
11:01
while a nearby loudspeaker played
183
661729
1741
ื‘ืขื•ื“ ืจืžืงื•ืœ ืงืจื•ื‘ ื ื™ื’ืŸ
11:03
the same "Mary Had a Little Lamb" music from before,
184
663470
2972
ืืช ืื•ืชื” ืžื•ื–ื™ืงื” ืฉืœ "ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ" ื›ืžื• ืžืงื•ื“ื,
11:06
but this time, we used just a regular store-bought camera,
185
666442
4203
ืื‘ืœ ื”ืคืขื, ื”ืฉืชืžืฉื ื• ื‘ืžืฆืœืžื” ืจื’ื™ืœื” ืฉื ืงื ืชื” ื‘ื—ื ื•ืช,
11:10
and so in a second, I'll play for you the sound that we recovered,
186
670645
3174
ื•ื›ืš ืขื•ื“ ืจื’ืข, ืื ื™ ืืฉืžื™ืข ืœื›ื ืืช ื”ืฆืœื™ืœ ืฉืฉื—ื–ืจื ื•,
11:13
and it's going to sound distorted this time,
187
673819
2050
ื•ื–ื” ืขื•ืžื“ ืœื”ืฉืžืข ืžืขื•ื•ืช ื”ืคืขื,
11:15
but listen and see if you can still recognize the music.
188
675869
2836
ืื‘ืœ ื”ืื–ื™ื ื• ื•ืชืจืื• ืื ืืชื ื™ื›ื•ืœื™ื ืขื“ื™ื™ืŸ ืœื–ื”ื•ืช ืืช ื”ืžื•ื–ื™ืงื”.
11:19
(Audio: "Mary Had a Little Lamb")
189
679723
6223
(ืื•ื“ื™ื•: ืœืžืจื™ ื”ื™ื” ื˜ืœื” ืงื˜ืŸ")
11:37
And so, again, that sounds distorted,
190
697527
3465
ื•ื›ืš, ืฉื•ื‘, ื–ื” ื ืฉืžืข ืžืขื•ื•ืช,
11:40
but what's really amazing here is that we were able to do this
191
700992
4386
ืื‘ืœ ืžื” ืฉื‘ืืžืช ืžื“ื”ื™ื ืคื” ื–ื” ืฉื”ื™ื™ื ื• ืžืกื•ื’ืœื™ื ืœืขืฉื•ืช ื–ืืช
11:45
with something that you could literally run out
192
705378
2626
ืขื ืžืฉื”ื• ืฉืืชื ื‘ืืžืช ื™ื›ื•ืœื™ื ืœืฆืืช
11:48
and pick up at a Best Buy.
193
708004
1444
ื•ืœืงื ื•ืช ื‘ื‘ืกื˜ ื‘ื™ื™.
11:51
So at this point,
194
711122
1363
ืื– ื‘ื ืงื•ื“ื” ื”ื–ื•,
11:52
a lot of people see this work,
195
712485
1974
ื”ืจื‘ื” ืื ืฉื™ื ืจื•ืื™ื ืืช ื”ืขื‘ื•ื“ื” ื”ื–ื•,
11:54
and they immediately think about surveillance.
196
714459
3413
ื•ื”ื ืžื™ื™ื“ ื—ื•ืฉื‘ื™ื ืขืœ ืžืขืงื‘.
11:57
And to be fair,
197
717872
2415
ื•ืœื”ื™ื•ืช ื›ื ื™ื,
12:00
it's not hard to imagine how you might use this technology to spy on someone.
198
720287
4133
ื–ื” ืœื ืงืฉื” ืœื“ืžื™ื™ืŸ ืื™ืš ืชื•ื›ืœื• ืœื”ืฉืชืžืฉ ื‘ื˜ื›ื ื•ืœื•ื’ื™ื” ื”ื–ื• ืœืจื’ืœ ืขืœ ืžื™ืฉื”ื•.
12:04
But keep in mind that there's already a lot of very mature technology
199
724420
3947
ืื‘ืœ ื–ื›ืจื• ืฉื™ืฉ ื›ื‘ืจ ื”ืจื‘ื” ื˜ื›ื ื•ืœื•ื’ื™ื•ืช ื‘ืฉืœื•ืช
12:08
out there for surveillance.
200
728367
1579
ืฉืงื™ื™ืžื•ืช ืœืžืขืงื‘.
12:09
In fact, people have been using lasers
201
729946
2090
ืœืžืขืฉื”, ืื ืฉื™ื ื”ืฉืชืžืฉื• ื‘ืœื™ื™ื–ืจื™ื
12:12
to eavesdrop on objects from a distance for decades.
202
732036
2799
ื›ื“ื™ ืœืฆื•ืชืช ืœืขืฆืžื™ื ืžืžืจื—ืง ื‘ืžืฉืš ืขืฉื•ืจื™ื.
12:15
But what's really new here,
203
735978
2025
ืื‘ืœ ืžื” ืฉื‘ืืžืช ื—ื“ืฉ ืคื”,
12:18
what's really different,
204
738003
1440
ืžื” ืฉื‘ืืžืช ืฉื•ื ื”,
12:19
is that now we have a way to picture the vibrations of an object,
205
739443
4295
ื–ื” ืฉืขื›ืฉื™ื• ื™ืฉ ืœื ื• ื“ืจืš ืœืฆืœื ืชืžื•ื ื•ืช ืฉืœ ืขืฆื,
12:23
which gives us a new lens through which to look at the world,
206
743738
3413
ืžื” ืฉื ื•ืชืŸ ืœื ื• ืขื“ืฉื” ื—ื“ืฉื” ื“ืจื›ื” ืœื”ืกืชื›ืœ ืขืœ ื”ืขื•ืœื,
12:27
and we can use that lens
207
747151
1510
ื•ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ืขื“ืฉื” ื”ื–ื•
12:28
to learn not just about forces like sound that cause an object to vibrate,
208
748661
4899
ื›ื“ื™ ืœืœืžื•ื“ ืœื ืจืง ืขืœ ื”ื›ื•ื—ื•ืช ื›ืžื• ืฆืœื™ืœ ืฉื’ื•ืจืžื™ื ืœืขืฆื ืœื ื•ืข,
12:33
but also about the object itself.
209
753560
2288
ืืœื ื’ื ืขืœ ื”ืขืฆื ืขืฆืžื•.
12:36
And so I want to take a step back
210
756975
1693
ืื– ืื ื™ ืจื•ืฆื” ืœืงื—ืช ืฆืขื“ ืื—ื•ืจื”
12:38
and think about how that might change the ways that we use video,
211
758668
4249
ื•ืœื—ืฉื•ื‘ ืขืœ ืื™ืš ื–ื” ืื•ืœื™ ื™ืฉื ื” ืืช ื”ื“ืจืš ื‘ื” ืื ื—ื ื• ืžืฉืชืžืฉื™ื ื‘ืกืจื˜ื•ื ื™ื,
12:42
because we usually use video to look at things,
212
762917
3553
ืžืคื ื™ ืฉืื ื—ื ื• ื‘ื“ืจืš ื›ืœืœ ืžืฉืชืžืฉื™ื ื‘ืกืจื˜ื•ื ื™ื ื›ื“ื™ ืœื”ื‘ื™ื˜ ื‘ื“ื‘ืจื™ื,
12:46
and I've just shown you how we can use it
213
766470
2322
ื•ืจืง ื”ืจืืชื™ ืœื›ื ืื™ืš ืื ื—ื ื• ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ื–ื”
12:48
to listen to things.
214
768792
1857
ื›ื“ื™ ืœื”ืงืฉื™ื‘ ืœื“ื‘ืจื™ื.
12:50
But there's another important way that we learn about the world:
215
770649
3971
ืื‘ืœ ื™ืฉ ื“ืจืš ื—ืฉื•ื‘ื” ื ื•ืกืคืช ืฉืื ื—ื ื• ื™ื›ื•ืœื™ื ืœืœืžื•ื“ ืขืœ ื”ืขื•ืœื:
12:54
that's by interacting with it.
216
774620
2275
ื–ื” ืขืœ ื™ื“ื™ ื”ืฉืคืขื” ืขืœื™ื•.
12:56
We push and pull and poke and prod things.
217
776895
3111
ืื ื—ื ื• ื“ื•ื—ืคื™ื ื•ืžื•ืฉื›ื™ื ื•ื“ื•ื—ืคื™ื ื•ืžืžืฉืฉื™ื ืขืฆืžื™ื.
13:00
We shake things and see what happens.
218
780006
3181
ืื—ื ื ื• ืžื ืขืจื™ื ื•ืจื•ืื™ื ืžื” ืงื•ืจื”.
13:03
And that's something that video still won't let us do,
219
783187
4273
ื•ื–ื” ืžืฉื”ื• ืฉื•ื™ื“ืื• ืขื“ื™ื™ืŸ ืœื ื ื•ืชืŸ ืœื ื• ืœืขืฉื•ืช,
13:07
at least not traditionally.
220
787460
2136
ืœืคื—ื•ืช ืœื ื‘ืื•ืคืŸ ืžืกื•ืจืชื™.
13:09
So I want to show you some new work,
221
789596
1950
ืื– ืจืฆื™ืชื™ ืœื”ืจืื•ืช ืœื›ื ืงืฆืช ืžื”ืขื‘ื•ื“ื” ื”ื—ื“ืฉื” ืฉืœื™,
13:11
and this is based on an idea I had just a few months ago,
222
791546
2667
ื•ื–ื” ืžื‘ื•ืกืก ืขืœ ืจืขื™ื•ืŸ ืฉื”ื™ื” ืœื™ ืจืง ืœืคื ื™ ื›ืžื” ื—ื•ื“ืฉื™ื,
13:14
so this is actually the first time I've shown it to a public audience.
223
794213
3301
ืื– ื–ืืช ืœืžืขืฉื” ื”ืคืขื ื”ืจืืฉื•ื ื” ืฉืื ื™ ืžืจืื” ืืช ื–ื” ืœืฆื™ื‘ื•ืจ.
13:17
And the basic idea is that we're going to use the vibrations in a video
224
797514
5363
ื•ื”ืจืขื™ื•ืŸ ื”ื‘ืกื™ืกื™ ื”ื•ื ืฉืื ื—ื ื• ืขื•ืžื“ื™ื ืœื”ืฉืชืžืฉ ื‘ืจืขื™ื“ื•ืช ื‘ืกืจื˜ื•ืŸ
13:22
to capture objects in a way that will let us interact with them
225
802877
4481
ื›ื“ื™ ืœืœื›ื•ื“ ืขืฆืžื™ื ื‘ื“ืจืš ืฉืชื™ืชืŸ ืœื ื• ืœื”ืฉืคื™ืข ืขืœื™ื”ื
13:27
and see how they react to us.
226
807358
1974
ื•ืœืจืื•ืช ืื™ืš ื”ื ืžื’ื™ื‘ื™ื ืœื ื•.
13:31
So here's an object,
227
811120
1764
ืื– ื”ื ื” ืขืฆื,
13:32
and in this case, it's a wire figure in the shape of a human,
228
812884
3832
ื•ื‘ืžืงืจื” ื”ื–ื”, ื–ื” ื“ืžื•ืช ื—ื•ื˜ ื‘ืฆื•ืจื” ืฉืœ ืื“ื,
13:36
and we're going to film that object with just a regular camera.
229
816716
3088
ื•ืื ื—ื ื• ื”ื•ืœื›ื™ื ืœืฆืœื ืืช ื”ืขืฆื ืขื ืžืฆืœืžื” ืจื’ื™ืœื”.
13:39
So there's nothing special about this camera.
230
819804
2124
ืื– ืื™ืŸ ืฉื•ื ื“ื‘ืจ ืžื™ื•ื—ื“ ื‘ืžืฆืœืžื” ื”ื–ื•.
13:41
In fact, I've actually done this with my cell phone before.
231
821928
2961
ืœืžืขืฉื”, ืขืฉื™ืชื™ ืืช ื–ื” ืขื ืžืฆืœืžืช ื”ืกืœื•ืœืจื™ ืฉืœื™ ืœืคื ื™ ื›ืŸ.
13:44
But we do want to see the object vibrate,
232
824889
2252
ืื‘ืœ ืื ื—ื ื• ืจื•ืฆื™ื ืœืจืื•ืช ืืช ื”ืื•ื‘ื™ื™ืงื˜ ืจื•ื˜ื˜.
13:47
so to make that happen,
233
827141
1133
ืื– ื›ื“ื™ ืœื’ืจื•ื ืœื–ื” ืœืงืจื•ืช,
13:48
we're just going to bang a little bit on the surface where it's resting
234
828274
3346
ืื ื—ื ื• ืคืฉื•ื˜ ื“ื•ืคืงื™ื ืžืขื˜ ืขืœ ื”ืžืฉื˜ื— ืขืœื™ื• ื”ื•ื ืžื•ื ื—
13:51
while we record this video.
235
831620
2138
ื‘ืขื•ื“ ืื ื—ื ื• ืžืงืœื™ื˜ื™ื ืืช ื”ืกืจื˜ื•ืŸ ื”ื–ื”.
13:59
So that's it: just five seconds of regular video,
236
839398
3671
ืื– ื–ื”ื•, ืจืง ื—ืžืฉ ืฉื ื™ื•ืช ืฉืœ ืกืจื˜ื•ืŸ ืจื’ื™ืœ,
14:03
while we bang on this surface,
237
843069
2136
ื‘ืขื•ื“ ืื ื—ื ื• ื“ื•ืคืงื™ื ืขืœ ื”ืžืฉื˜ื— ื”ื–ื”,
14:05
and we're going to use the vibrations in that video
238
845205
3513
ื•ืื ื—ื ื• ืขื•ืžื“ื™ื ืœื”ืฉืชืžืฉ ื‘ืจืขื™ื“ื•ืช ื‘ืกืจื˜ื•ืŸ
14:08
to learn about the structural and material properties of our object,
239
848718
4544
ื›ื“ื™ ืœืœืžื•ื“ ืขืœ ื”ืชื›ื•ื ื•ืช ื”ืžื‘ื ื™ื•ืช ื•ื”ื—ื•ืžืจื™ื•ืช ืฉืœ ื”ืขืฆื ืฉืœื ื•,
14:13
and we're going to use that information to create something new and interactive.
240
853262
4834
ื•ืื ื—ื ื• ื”ื•ืœื›ื™ื ืœื”ืฉืชืžืฉ ื‘ืžื™ื“ืข ื”ื–ื” ื›ื“ื™ ืœื™ืฆื•ืจ ืžืฉื”ื• ื—ื“ืฉ ื•ืื™ื ื˜ืจืืงื˜ื™ื‘ื™.
14:24
And so here's what we've created.
241
864866
2653
ื•ื”ื ื” ืžื” ืฉื™ืฆืจื ื•.
14:27
And it looks like a regular image,
242
867519
2229
ื•ื–ื” ื ืจืื” ื›ืžื• ืชืžื•ื ื” ืจื’ื™ืœื”,
14:29
but this isn't an image, and it's not a video,
243
869748
3111
ืื‘ืœ ื–ื• ืœื ืชืžื•ื ื”, ื•ื–ื” ืœื ืกืจื˜ื•ืŸ,
14:32
because now I can take my mouse
244
872859
2368
ืžืคื ื™ ืฉืขื›ืฉื™ื• ืื ื™ ื™ื›ื•ืœ ืœืงื—ืช ืืช ื”ืขื›ื‘ืจ ืฉืœื™
14:35
and I can start interacting with the object.
245
875227
2859
ื•ืื ื™ ื™ื›ื•ืœ ืœื”ืชื—ื™ืœ ืœื”ืฉืคื™ืข ืขืœ ื”ืขืฆื.
14:44
And so what you see here
246
884936
2357
ื•ื›ืš ืžื” ืฉืืชื ืจื•ืื™ื ืคื”
14:47
is a simulation of how this object
247
887389
2226
ื–ื” ื”ื“ืžื™ื™ื” ืฉืœ ืื™ืš ื”ืขืฆื ื”ื–ื”
14:49
would respond to new forces that we've never seen before,
248
889615
4458
ื™ื’ื™ื‘ ืœื›ื•ื—ื•ืช ื—ื“ืฉื™ื ืฉืžืขื•ืœื ืœื ืจืื™ื ื• ืœืคื ื™ ื›ืŸ,
14:54
and we created it from just five seconds of regular video.
249
894073
3633
ื•ื™ืฆืจื ื• ืืช ื–ื” ืจืง ืžื—ืžืฉ ืฉื ื™ื•ืช ืฉืœ ืกืจื˜ื•ืŸ ืจื’ื™ืœ.
14:59
(Applause)
250
899249
4715
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
15:09
And so this is a really powerful way to look at the world,
251
909421
3227
ื•ื–ื• ื“ืจืš ื‘ืืžืช ื—ื–ืงื” ืœื”ื‘ื™ื˜ ื‘ืขื•ืœื,
15:12
because it lets us predict how objects will respond
252
912648
2972
ืžืคื ื™ ืฉื”ื™ื ืžืืคืฉืจืช ืœื ื• ืœื—ื–ื•ืช ืื™ืš ืขืฆืžื™ื ื™ื’ื™ื‘ื•
15:15
to new situations,
253
915620
1823
ืœืžืฆื‘ื™ื ื—ื“ืฉื™ื,
15:17
and you could imagine, for instance, looking at an old bridge
254
917443
3473
ื•ืืชื ื™ื›ื•ืœื™ื ืœื“ืžื™ื™ืŸ, ืœื“ื•ื’ืžื”, ืฉืชื‘ื™ื˜ื• ื‘ื’ืฉืจ ื™ืฉืŸ
15:20
and wondering what would happen, how would that bridge hold up
255
920916
3527
ื•ืชืชื”ื• ืžื” ื™ืงืจื”, ืื™ืš ื”ื’ืฉืจ ื™ื—ื–ื™ืง
15:24
if I were to drive my car across it.
256
924443
2833
ืื ืžื›ื•ื ื™ืช ืชื™ืกืข ืขืœื™ื•.
15:27
And that's a question that you probably want to answer
257
927276
2774
ื•ื–ื• ืฉืืœื” ืฉืืชื ื›ื ืจืื” ืจื•ืฆื™ื ืœืขื ื•ืช ืขืœื™ื”
15:30
before you start driving across that bridge.
258
930050
2560
ืœืคื ื™ ืฉืืชื ืžืชื—ื™ืœื™ื ืœื ืกื•ืข ืœืื•ืจืš ื”ื’ืฉืจ.
15:33
And of course, there are going to be limitations to this technique,
259
933988
3272
ื•ื›ืžื•ื‘ืŸ, ื™ื”ื™ื• ืžื’ื‘ืœื•ืช ืขืœ ื”ืฉื™ื˜ื” ื”ื–ื•,
15:37
just like there were with the visual microphone,
260
937260
2462
ืžืžืฉ ื›ืžื• ืฉื™ืฉ ืขื ื”ืžื™ืงืจื•ืคื•ืŸ ื”ื•ื™ื–ื•ืืœื™,
15:39
but we found that it works in a lot of situations
261
939722
3181
ืื‘ืœ ื’ื™ืœื™ื ื• ืฉื–ื” ืขื•ื‘ื“ ื‘ื”ืจื‘ื” ืžืงืจื™ื
15:42
that you might not expect,
262
942903
1875
ืฉืื•ืœื™ ืœื ื”ื™ื™ืชื ืžืฆืคื™ื,
15:44
especially if you give it longer videos.
263
944778
2768
ื‘ืขื™ืงืจ ืื ืืชื ื ื•ืชื ื™ื ืœื–ื” ืกืจื˜ื•ืŸ ืืจื•ืš ื™ื•ืชืจ.
15:47
So for example, here's a video that I captured
264
947546
2508
ืื– ืœื“ื•ื’ืžื”, ื”ื ื” ืกืจื˜ื•ืŸ ืฉืฆื™ืœืžื ื•
15:50
of a bush outside of my apartment,
265
950054
2299
ืฉืœ ืฉื™ื— ืžื—ื•ืฅ ืœื“ื™ืจื” ืฉืœื™,
15:52
and I didn't do anything to this bush,
266
952353
3088
ื•ืœื ืขืฉื™ืชื™ ื›ืœื•ื ืœืฉื™ื— ื”ื–ื”,
15:55
but by capturing a minute-long video,
267
955441
2705
ืื‘ืœ ืขืœ ื™ื“ื™ ืฆื™ืœื•ื ืฉืœ ืกืจื˜ื•ืŸ ื‘ืื•ืจืš ืฉืœ ื“ืงื”,
15:58
a gentle breeze caused enough vibrations
268
958146
3378
ืจื•ื— ืงืœื” ื’ืจืžื” ืœืžืกืคื™ืง ืจืขื™ื“ื•ืช
16:01
that we could learn enough about this bush to create this simulation.
269
961524
3587
ืฉื™ื›ื•ืœื ื• ืœืœืžื•ื“ ืžืกืคื™ืง ืขืœ ื”ืฉื™ื— ื”ื–ื” ื›ื“ื™ ืœื™ืฆื•ืจ ืืช ื”ื“ืžื™ื™ื” ื”ื–ื•.
16:07
(Applause)
270
967270
6142
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
16:13
And so you could imagine giving this to a film director,
271
973412
2972
ื•ื›ืš ืชื•ื›ืœื• ืœื“ืžื™ื™ื™ืŸ ืฉืชืชื ื• ืืช ื–ื” ืœื‘ืžืื™ ืกืจื˜ื™ื,
16:16
and letting him control, say,
272
976384
1719
ื•ืชืชื ื• ืœื• ืฉืœื™ื˜ื”, ื ื’ื™ื“,
16:18
the strength and direction of wind in a shot after it's been recorded.
273
978103
4922
ื‘ืขื•ืฆืžื” ื•ื‘ื›ื™ื•ื•ืŸ ืฉืœ ื”ืจื•ื— ื‘ืฆื™ืœื•ื ืœืื—ืจ ืฉื”ื•ื ื”ื•ืงืœื˜.
16:24
Or, in this case, we pointed our camera at a hanging curtain,
274
984810
4535
ืื•, ื‘ืžืงืจื” ื”ื–ื”, ื›ื™ื•ื•ื ื• ืืช ื”ืžืฆืœืžื” ืœื•ื™ืœื•ืŸ ืชืœื•ื™,
16:29
and you can't even see any motion in this video,
275
989345
4129
ื•ืืชื ืœื ื™ื›ื•ืœื™ื ืืคื™ืœื• ืœืจืื•ืช ืชื ื•ืขื” ื‘ืกืจื˜ื•ืŸ ื”ื–ื”,
16:33
but by recording a two-minute-long video,
276
993474
2925
ืื‘ืœ ืขืœ ื™ื“ื™ ื”ืงืœื˜ืช ืกืจื˜ื•ืŸ ืฉืœ ืฉืชื™ ื“ืงื•ืช,
16:36
natural air currents in this room
277
996399
2438
ื–ืจืžื™ ืื•ื™ืจ ื˜ื‘ืขื™ื™ื ื‘ื—ื“ืจ ื”ื–ื”
16:38
created enough subtle, imperceptible motions and vibrations
278
998837
4412
ื™ืฆืจื• ืžืกืคื™ืง ืชื ื•ืขื•ืช ื•ืจืขื™ื“ื•ืช ืขื“ื™ื ื•ืช ื•ื‘ืœืชื™ ื ืงืœื˜ื•ืช
16:43
that we could learn enough to create this simulation.
279
1003249
2565
ืฉื™ื›ื•ืœื ื• ืœืœืžื•ื“ ืžืกืคื™ืง ื›ื“ื™ ืœื™ืฆื•ืจ ืืช ื”ื”ื“ืžื™ื” ื”ื–ื•.
16:48
And ironically,
280
1008243
2366
ื•ื‘ืื•ืคืŸ ืื™ืจื•ื ื™,
16:50
we're kind of used to having this kind of interactivity
281
1010609
3088
ืื ื—ื ื• ื“ื™ ืจื’ื™ืœื™ื ืฉื™ืฉ ืœื ื• ืืช ื™ื›ื•ืœืช ื”ื”ืฉืคืขื” ื”ื–ื•
16:53
when it comes to virtual objects,
282
1013697
2647
ื›ืฉื–ื” ืžื’ื™ืข ืœืขืฆืžื™ื ื•ื™ืจื˜ื•ืืœื™ื,
16:56
when it comes to video games and 3D models,
283
1016344
3297
ื›ืฉื–ื” ืžื’ื™ืข ืœืžืฉื—ืงื™ ืžื—ืฉื‘ ื•ืžื•ื“ืœื™ื ืชืœืช ืžื™ืžื“ื™ื™ื,
16:59
but to be able to capture this information from real objects in the real world
284
1019641
4404
ืื‘ืœ ื›ื“ื™ ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื ืœืœื›ื•ื“ ืืช ื”ืžื™ื“ืข ื”ื–ื” ืžืขืฆืžื™ื ืืžื™ืชื™ื™ื ื‘ืขื•ืœื ื”ืืžื™ืชื™
17:04
using just simple, regular video,
285
1024045
2817
ื‘ืฉื™ืžื•ืฉ ื‘ืกืจื˜ื•ืŸ ืคืฉื•ื˜ ื•ืจื’ื™ืœ,
17:06
is something new that has a lot of potential.
286
1026862
2183
ื–ื” ืžืฉื”ื• ื—ื“ืฉ ืฉื™ืฉ ืœื• ืคื•ื˜ื ืฆื™ืืœ ื’ื“ื•ืœ.
17:10
So here are the amazing people who worked with me on these projects.
287
1030410
4904
ืื– ื”ื ื” ื”ืื ืฉื™ื ื”ืžื“ื”ื™ืžื™ื ืฉืขื‘ื“ื• ืื™ืชื™ ืขืœ ื”ืคืจื•ื™ื™ืงื˜ื™ื ื”ืืœื”.
17:16
(Applause)
288
1036057
5596
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
17:24
And what I've shown you today is only the beginning.
289
1044819
3057
ื•ืžื” ืฉื”ืจืืชื™ ืœื›ื ื”ื™ื•ื ื–ื• ืจืง ื”ื”ืชื—ืœื”.
17:27
We've just started to scratch the surface
290
1047876
2113
ืจืง ื”ืชื—ืœื ื• ืœื’ืจื“ ืืช ืคื ื™ ื”ืฉื˜ื—
17:29
of what you can do with this kind of imaging,
291
1049989
2972
ืฉืœ ืžื” ืืคืฉืจ ืœืขืฉื•ืช ืขื ืชืžื•ื ื•ืช ืžืกื•ื’ ื–ื”,
17:32
because it gives us a new way
292
1052961
2286
ืžืคื ื™ ืฉื–ื” ื ื•ืชืŸ ืœื ื• ื“ืจืš ื—ื“ืฉื”
17:35
to capture our surroundings with common, accessible technology.
293
1055342
4724
ืœืœื›ื•ื“ ืืช ื”ืกื‘ื™ื‘ื” ืฉืœื ื• ืขื ื˜ื›ื ื•ืœื•ื’ื™ื” ื ืคื•ืฆื” ื•ื ื’ื™ืฉื”.
17:40
And so looking to the future,
294
1060066
1929
ื•ื‘ืฆื™ืคื™ื” ืœืขืชื™ื“,
17:41
it's going to be really exciting to explore
295
1061995
2037
ื–ื” ืขื•ืžื“ ืœื”ื™ื•ืช ืžืื•ื“ ืžืจื’ืฉ ืœื—ืงื•ืจ
17:44
what this can tell us about the world.
296
1064032
1856
ืžื” ื–ื” ื™ื›ื•ืœ ืœื”ื’ื™ื“ ืœื ื• ืขืœ ื”ืขื•ืœื.
17:46
Thank you.
297
1066381
1204
ืชื•ื“ื” ืœื›ื.
17:47
(Applause)
298
1067610
6107
(ืžื—ื™ืืช ื›ืคื™ื™ื)
ื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