TED2015
Abe Davis: New video technology that reveals an object's hidden properties
Ejb Dejvis (Abe Davis): Nova video tehnologija koja otkriva skrivena svojstva predmeta
Filmed:
Readability: 4.3
1,482,525 views
Jedva primetna kretanja dešavaju se stalno oko nas, uključujući sitne vibracije koje izaziva zvuk. Nova tehnologija pokazuje da možemo čuti ove vibracije i ponovo stvoriti zvuk i razgovore iz prostog video zapisa naoko nepomičnog predmeta. Međutim, Ejb Dejvis sada ide korak dalje: pogledajte kako prikazuje softver koji omogućava svakom da se igra ovim skrivenim svojstvima, koristeći samo prost video zapis.
Abe Davis - Computer scientist
Computer vision expert Abe Davis pioneers methods to extract audio from silent digital videos, even footage shot on ordinary consumer cameras. Full bio
Computer vision expert Abe Davis pioneers methods to extract audio from silent digital videos, even footage shot on ordinary consumer cameras. Full bio
Double-click the English transcript below to play the video.
00:13
Most of us think of motion
as a very visual thing.
as a very visual thing.
0
1373
3349
Većina nas shvata pokret
kao nešto izrazito vizuelno.
kao nešto izrazito vizuelno.
00:17
If I walk across this stage
or gesture with my hands while I speak,
or gesture with my hands while I speak,
1
5889
5088
Ako hodam po ovoj bini
ili mrdam rukama dok govorim,
ili mrdam rukama dok govorim,
00:22
that motion is something that you can see.
2
10977
2261
taj pokret je nešto što možete
da vidite.
da vidite.
00:26
But there's a world of important motion
that's too subtle for the human eye,
that's too subtle for the human eye,
3
14255
5482
Ali postoji ceo svet važnih pokreta
koji je nevidljiv za ljudsko oko,
koji je nevidljiv za ljudsko oko,
00:31
and over the past few years,
4
19737
2041
i proteklih nekoliko godina,
00:33
we've started to find that cameras
5
21778
1997
počeo sam da shvatam da kamere
00:35
can often see this motion
even when humans can't.
even when humans can't.
6
23775
3410
često mogu da vide pokret
koji ljudsko oko ne može.
koji ljudsko oko ne može.
00:40
So let me show you what I mean.
7
28305
1551
Pokazaću vam na šta mislim.
00:42
On the left here, you see video
of a person's wrist,
of a person's wrist,
8
30717
3622
Sa leve strane, možete videti snimak
nečijeg zgloba,
nečijeg zgloba,
00:46
and on the right, you see video
of a sleeping infant,
of a sleeping infant,
9
34339
3147
a sa desne strane, snimak bebe koja spava,
00:49
but if I didn't tell you
that these were videos,
that these were videos,
10
37486
3146
ali da vam nisam rekao
da su ovo video snimci,
da su ovo video snimci,
00:52
you might assume that you were looking
at two regular images,
at two regular images,
11
40632
3761
možda biste pomislili da gledate
u dve najobičnije slike,
u dve najobičnije slike,
00:56
because in both cases,
12
44393
1672
zato što u oba slučaja,
00:58
these videos appear to be
almost completely still.
almost completely still.
13
46065
3047
ovi snimci deluju potpuno mirno.
01:02
But there's actually a lot
of subtle motion going on here,
of subtle motion going on here,
14
50175
3885
Zapravo, dosta nevidljivih pokreta
imamo ovde,
imamo ovde,
01:06
and if you were to touch
the wrist on the left,
the wrist on the left,
15
54060
2392
i ako biste dotakli zglob na levoj strani,
01:08
you would feel a pulse,
16
56452
1996
osetili biste puls,
01:10
and if you were to hold
the infant on the right,
the infant on the right,
17
58448
2485
a kada biste držali bebu, s desne strane,
01:12
you would feel the rise
and fall of her chest
and fall of her chest
18
60933
2391
ostetili biste podizanje
i spuštanje njenih grudi
i spuštanje njenih grudi
01:15
as she took each breath.
19
63324
1390
dok udiše i izdiše.
01:17
And these motions carry
a lot of significance,
a lot of significance,
20
65762
3576
Ovi pokreti su od velikog značaja,
01:21
but they're usually
too subtle for us to see,
too subtle for us to see,
21
69338
3343
ali su obično jako mali
da bi ih mi uočili,
da bi ih mi uočili,
01:24
so instead, we have to observe them
22
72681
2276
pa zbog toga, moramo da ih posmatramo
01:26
through direct contact, through touch.
23
74957
2900
kroz direktni kontakt, kroz dodir.
01:30
But a few years ago,
24
78997
1265
Ali pre nekoliko godina,
01:32
my colleagues at MIT developed
what they call a motion microscope,
what they call a motion microscope,
25
80262
4405
moj kolega sa MIT-a je razvio
nešto što oni zovu mikroskopom pokreta,
nešto što oni zovu mikroskopom pokreta,
01:36
which is software that finds
these subtle motions in video
these subtle motions in video
26
84667
4384
što je zapravo softver koji nalazi
ove male pokrete, snima ih
ove male pokrete, snima ih
01:41
and amplifies them so that they
become large enough for us to see.
become large enough for us to see.
27
89051
3562
i uvećava pa time postaju dovoljno
veliki da ih mi vidimo.
veliki da ih mi vidimo.
01:45
And so, if we use their software
on the left video,
on the left video,
28
93416
3483
Dakle, ako primenimo softver
na levi snimak,
na levi snimak,
01:48
it lets us see the pulse in this wrist,
29
96899
3250
omogućava nam da vidimo
puls na zglobu,
puls na zglobu,
01:52
and if we were to count that pulse,
30
100149
1695
a ako bismo merili puls,
01:53
we could even figure out
this person's heart rate.
this person's heart rate.
31
101844
2355
mogli bismo čak i izmeriti
otkucaj srca ove osobe.
otkucaj srca ove osobe.
01:57
And if we used the same software
on the right video,
on the right video,
32
105095
3065
A ako primenimo isti softver
na snimak desno,
na snimak desno,
02:00
it lets us see each breath
that this infant takes,
that this infant takes,
33
108160
3227
moći ćemo da vidimo svaki udisaj bebe,
02:03
and we can use this as a contact-free way
to monitor her breathing.
to monitor her breathing.
34
111387
4137
i možemo ovo da iskoristimo kao nedirektno
nadgledanje njenog disanja.
nadgledanje njenog disanja.
02:08
And so this technology is really powerful
because it takes these phenomena
because it takes these phenomena
35
116884
5348
Dakle, ova tehnologija je jako moćna,
zato što ove pojave
zato što ove pojave
02:14
that we normally have
to experience through touch
to experience through touch
36
122232
2367
u normalnim uslovima
doživljavamo kroz dodir,
doživljavamo kroz dodir,
02:16
and it lets us capture them visually
and non-invasively.
and non-invasively.
37
124599
2957
i pamtimo ih vizuelno i neinvazivno.
02:21
So a couple years ago, I started working
with the folks that created that software,
with the folks that created that software,
38
129104
4411
Pre par godina, počeo sam da radim
sa ljudima koji su napravili taj softver,
sa ljudima koji su napravili taj softver,
02:25
and we decided to pursue a crazy idea.
39
133515
3367
i odlučili smo da pratimo
jednu ludu ideju.
jednu ludu ideju.
02:28
We thought, it's cool
that we can use software
that we can use software
40
136882
2693
Pomislili smo, kul je što
možemo da upotrebimo softver
možemo da upotrebimo softver
02:31
to visualize tiny motions like this,
41
139575
3135
da predočimo majušne pokrete
poput ovog,
poput ovog,
02:34
and you can almost think of it
as a way to extend our sense of touch.
as a way to extend our sense of touch.
42
142710
4458
a možemo to shavatiti kao način
kojim pojačavamo naš osećaj za dodir.
kojim pojačavamo naš osećaj za dodir.
02:39
But what if we could do the same thing
with our ability to hear?
with our ability to hear?
43
147168
4059
Ali šta ako bismo mogli da učinimo
isto sa osećajem sluha?
isto sa osećajem sluha?
02:44
What if we could use video
to capture the vibrations of sound,
to capture the vibrations of sound,
44
152508
4665
Šta ako bismo mogli da koristimo video
za beleženje zvučnih vibracija,
za beleženje zvučnih vibracija,
02:49
which are just another kind of motion,
45
157173
2827
koje su samo druga vrsta pokreta,
02:52
and turn everything that we see
into a microphone?
into a microphone?
46
160000
3346
i pretvorimo sve što vidimo u mikrofon?
02:56
Now, this is a bit of a strange idea,
47
164236
1971
Sad, ovo je pomalo čudna ideja,
02:58
so let me try to put it
in perspective for you.
in perspective for you.
48
166207
2586
pa me pustite da vam predočim.
03:01
Traditional microphones
work by converting the motion
work by converting the motion
49
169523
3488
Obični mikrofoni funkcionišu tako što
pretvaraju vibriranje
pretvaraju vibriranje
03:05
of an internal diaphragm
into an electrical signal,
into an electrical signal,
50
173011
3599
unutrašnje dijafragme
u elektornski signal,
u elektornski signal,
03:08
and that diaphragm is designed
to move readily with sound
to move readily with sound
51
176610
4318
a ta dijafragma je dizajnirana
da se pomera uporedo sa zvukom,
da se pomera uporedo sa zvukom,
03:12
so that its motion can be recorded
and interpreted as audio.
and interpreted as audio.
52
180928
4807
pa se njeno vibriranje beleži
i interpretira kao audio zapis.
i interpretira kao audio zapis.
03:17
But sound causes all objects to vibrate.
53
185735
3668
Međutim, zvuk uzrokuje
da svi objekti vibriraju.
da svi objekti vibriraju.
03:21
Those vibrations are just usually
too subtle and too fast for us to see.
too subtle and too fast for us to see.
54
189403
5480
A te vibracije su obično neprimetne
i jako brze za nas da bismo ih uočili.
i jako brze za nas da bismo ih uočili.
03:26
So what if we record them
with a high-speed camera
with a high-speed camera
55
194883
3738
Dakle, šta ako bi ih beležili
kamerom velike brzine,
kamerom velike brzine,
03:30
and then use software
to extract tiny motions
to extract tiny motions
56
198621
3576
a zatim upotrebili softver
da pojačamo sićušne vibracije
da pojačamo sićušne vibracije
03:34
from our high-speed video,
57
202197
2090
sa našeg brzog snimka,
03:36
and analyze those motions to figure out
what sounds created them?
what sounds created them?
58
204287
4274
i analiziramo te vibracije
da bismo shvatili kakvi ih zvuci stvaraju?
da bismo shvatili kakvi ih zvuci stvaraju?
03:41
This would let us turn visible objects
into visual microphones from a distance.
into visual microphones from a distance.
59
209859
5449
Ovo bi nam dozvolilo da prebacimo vidljive
predmete u vizuelne mikrofone sa distance.
predmete u vizuelne mikrofone sa distance.
03:49
And so we tried this out,
60
217080
2183
Pa smo ovo i isprobali.
03:51
and here's one of our experiments,
61
219263
1927
Evo jednog od eksperimenata,
gde smo uzeli ovu biljku u saksiji,
koju možete da vidite sa desne strane,
koju možete da vidite sa desne strane,
03:53
where we took this potted plant
that you see on the right
that you see on the right
62
221190
2949
03:56
and we filmed it with a high-speed camera
63
224139
2438
i snimili smo kamerom velike brzine,
03:58
while a nearby loudspeaker
played this sound.
played this sound.
64
226577
3529
dok je zvučnik u blizini puštao ovaj zvuk.
04:02
(Music: "Mary Had a Little Lamb")
65
230275
8190
(muzika: "Meri je imala malo jagnje")
04:11
And so here's the video that we recorded,
66
239820
2824
A evo snimka koji smo snimili,
04:14
and we recorded it at thousands
of frames per second,
of frames per second,
67
242644
3924
a snimili smo ga
pri 1000 sličica u sekundi,
pri 1000 sličica u sekundi,
04:18
but even if you look very closely,
68
246568
2322
ali čak i kada pogledate bliže,
04:20
all you'll see are some leaves
69
248890
1951
sve što vidite su neki listovi
04:22
that are pretty much
just sitting there doing nothing,
just sitting there doing nothing,
70
250841
3065
koji jednostavno samo stoje
i ne rade bilo šta,
i ne rade bilo šta,
04:25
because our sound only moved those leaves
by about a micrometer.
by about a micrometer.
71
253906
4806
zato što je naš zvuk pomerao ove listove
samo za mikrometar ili dva.
samo za mikrometar ili dva.
04:31
That's one ten-thousandth of a centimeter,
72
259103
4276
To je desetohiljaditi deo centimetra,
04:35
which spans somewhere between
a hundredth and a thousandth
a hundredth and a thousandth
73
263379
4156
što predstavlja između stotog
i hiljaditog dela piksela
i hiljaditog dela piksela
04:39
of a pixel in this image.
74
267535
2299
na ovoj slici.
04:41
So you can squint all you want,
75
269881
2887
Dakle, možete škiljiti koliko god želite,
04:44
but motion that small is pretty much
perceptually invisible.
perceptually invisible.
76
272768
3335
ali toliko mali pokret je vizuelno
prilično nevidljiv.
prilično nevidljiv.
04:49
But it turns out that something
can be perceptually invisible
can be perceptually invisible
77
277667
4157
Ali ispostavlja se da nešto što može
da bude čulno nevidljivo
da bude čulno nevidljivo
04:53
and still be numerically significant,
78
281824
2809
i dalje može biti numerički značajno,
04:56
because with the right algorithms,
79
284633
2002
time što sa adekvatnim algoritmima,
04:58
we can take this silent,
seemingly still video
seemingly still video
80
286635
3687
možemo uzeti ovaj tih,
naizgled miran snimak
naizgled miran snimak
05:02
and we can recover this sound.
81
290322
1527
i vratimo ovaj zvuk.
05:04
(Music: "Mary Had a Little Lamb")
82
292690
7384
(muzika: "Meri je imala malo jagnje")
05:12
(Applause)
83
300074
5828
(Aplauz)
05:22
So how is this possible?
84
310058
1939
Kako je ovo moguće?
05:23
How can we get so much information
out of so little motion?
out of so little motion?
85
311997
4344
Kako možemo dobiti toliko informacija
iz toliko sitnih pokreta?
iz toliko sitnih pokreta?
05:28
Well, let's say that those leaves
move by just a single micrometer,
move by just a single micrometer,
86
316341
5361
Pa, recimo da se ti listovi pomeraju
za jedan mikrometar,
za jedan mikrometar,
05:33
and let's say that that shifts our image
by just a thousandth of a pixel.
by just a thousandth of a pixel.
87
321702
4308
i recomo da to pomera našu sliku
za jedan hiljaditi deo piksela.
za jedan hiljaditi deo piksela.
05:39
That may not seem like much,
88
327269
2572
To možda ne deluje mnogo,
05:41
but a single frame of video
89
329841
1996
ali jedna sličica snimka
05:43
may have hundreds of thousands
of pixels in it,
of pixels in it,
90
331837
3257
može da sadrži stotine hiljada
piksela u sebi,
piksela u sebi,
05:47
and so if we combine all
of the tiny motions that we see
of the tiny motions that we see
91
335094
3454
pa tako ako iskombinujemo
sve ove sitne pokrete koje vidimo
sve ove sitne pokrete koje vidimo
05:50
from across that entire image,
92
338548
2298
preko cele slike,
05:52
then suddenly a thousandth of a pixel
93
340846
2623
odjednom hiljaditi deo piksela
05:55
can start to add up
to something pretty significant.
to something pretty significant.
94
343469
2775
može doprineti nečemu jako bitnom.
05:58
On a personal note, we were pretty psyched
when we figured this out.
when we figured this out.
95
346870
3635
Lično, bili smo prilično zapanjeni
kada smo ovo otkrili.
kada smo ovo otkrili.
06:02
(Laughter)
96
350505
2320
(Smeh)
06:04
But even with the right algorithm,
97
352825
3253
Ali i sa pravim algoritmom,
06:08
we were still missing
a pretty important piece of the puzzle.
a pretty important piece of the puzzle.
98
356078
3617
i dalje nam je nedostajao
jako bitan deo slagalice.
jako bitan deo slagalice.
06:11
You see, there are a lot of factors
that affect when and how well
that affect when and how well
99
359695
3604
Vidite, postoji veliki broj faktora
koji utiču kada i koliko dobro će
koji utiču kada i koliko dobro će
06:15
this technique will work.
100
363299
1997
ova tehika raditi.
06:17
There's the object and how far away it is;
101
365296
3204
U pitanju je objekat
i koliko je on daleko;
i koliko je on daleko;
06:20
there's the camera
and the lens that you use;
and the lens that you use;
102
368500
2394
zatim, kamera i sočiva koja koristite;
06:22
how much light is shining on the object
and how loud your sound is.
and how loud your sound is.
103
370894
4091
koliko je objekat osvetljen
i koliko je jak vaš zvuk.
i koliko je jak vaš zvuk.
06:27
And even with the right algorithm,
104
375945
3375
I sa odgovarajućim algoritmom,
06:31
we had to be very careful
with our early experiments,
with our early experiments,
105
379320
3390
moramo biti jako pažljivi
sa našim prvim eksperimentima,
sa našim prvim eksperimentima,
zato što ako bilo koji od
ovih faktora bude poremećen,
ovih faktora bude poremećen,
06:34
because if we got
any of these factors wrong,
any of these factors wrong,
106
382710
2392
06:37
there was no way to tell
what the problem was.
what the problem was.
107
385102
2368
ne postoji način da shvatimo
šta je u pitanju.
šta je u pitanju.
06:39
We would just get noise back.
108
387470
2647
Zauzvrat bismo dobili samo buku.
06:42
And so a lot of our early
experiments looked like this.
experiments looked like this.
109
390117
3320
Pa prema tome, većina naših
prvih eksperimenata je ovako izgledala.
prvih eksperimenata je ovako izgledala.
06:45
And so here I am,
110
393437
2206
Evo mene,
06:47
and on the bottom left, you can kind of
see our high-speed camera,
see our high-speed camera,
111
395643
4040
dole levo možete videti kameru
velike brzine,
velike brzine,
06:51
which is pointed at a bag of chips,
112
399683
2183
koja je usmerena ka kesici čipsa,
06:53
and the whole thing is lit
by these bright lamps.
by these bright lamps.
113
401866
2949
a cela stvar je osvetljena
ovim jakim lampama.
ovim jakim lampama.
06:56
And like I said, we had to be
very careful in these early experiments,
very careful in these early experiments,
114
404815
4365
Kao što sam rekao, morali smo da budemo
jako pažljivi sa prvim eksperimentima,
jako pažljivi sa prvim eksperimentima,
07:01
so this is how it went down.
115
409180
2508
i to je ovako izgledalo.
07:03
(Video) Abe Davis: Three, two, one, go.
116
411688
3761
(Snimak) Eb Dejvis:
Tri, dva, jedan, kreni.
Tri, dva, jedan, kreni.
07:07
Mary had a little lamb!
Little lamb! Little lamb!
Little lamb! Little lamb!
117
415449
5387
Meri je imala malo jagnje!
Malo jagnje! Malo jagnje!
Malo jagnje! Malo jagnje!
07:12
(Laughter)
118
420836
4500
(Smeh)
07:17
AD: So this experiment
looks completely ridiculous.
looks completely ridiculous.
119
425336
2814
ED: Dakle, ovaj eksperiment
je izgledao potpuno smešno.
je izgledao potpuno smešno.
07:20
(Laughter)
120
428150
1788
(Smeh)
07:21
I mean, I'm screaming at a bag of chips --
121
429938
2345
Mislim, vičem na kesicu čipsa,
07:24
(Laughter) --
122
432283
1551
(Smeh)
07:25
and we're blasting it with so much light,
123
433834
2117
i osvetili smo je toliko jakom svetlošću,
07:27
we literally melted the first bag
we tried this on. (Laughter)
we tried this on. (Laughter)
124
435951
4479
da smo bukvalno istopili prvu kesicu
kada smo probali ovo. (Smeh)
kada smo probali ovo. (Smeh)
07:32
But ridiculous as this experiment looks,
125
440525
3274
Ali koliko god smešno izgledao
ovaj eksperiment,
ovaj eksperiment,
07:35
it was actually really important,
126
443799
1788
bio je, zapravo, jako bitan,
07:37
because we were able
to recover this sound.
to recover this sound.
127
445587
2926
zato što smo mogli da povratimo zvuk.
07:40
(Audio) Mary had a little lamb!
Little lamb! Little lamb!
Little lamb! Little lamb!
128
448513
4712
(Audio) Meri je imala malo jagnje!
Malo jagnje! Malo jagnje!
Malo jagnje! Malo jagnje!
07:45
(Applause)
129
453225
4088
(Aplauz)
07:49
AD: And this was really significant,
130
457313
1881
ED: Ovo je bilo značajno,
07:51
because it was the first time
we recovered intelligible human speech
we recovered intelligible human speech
131
459194
4119
zato što je to bio prvi put da smo uspeli
da povratimo razumljiv ljudski govor
da povratimo razumljiv ljudski govor
07:55
from silent video of an object.
132
463424
2341
tihog snimka jednog objekta.
07:57
And so it gave us this point of reference,
133
465765
2391
Dakle, ovo nam je dalo ovakve rezultate,
08:00
and gradually we could start
to modify the experiment,
to modify the experiment,
134
468156
3871
i vremenom smo mogli da počnemo
da modifikujemo eksperiment
da modifikujemo eksperiment
08:04
using different objects
or moving the object further away,
or moving the object further away,
135
472106
3805
koristeći različite objekte
ili pomerajući objekat dalje,
ili pomerajući objekat dalje,
08:07
using less light or quieter sounds.
136
475911
2770
koristeći slabije svetlo ili tiši zvuk.
08:11
And we analyzed all of these experiments
137
479887
2874
Analizirali smo sve ove ekperimente
08:14
until we really understood
the limits of our technique,
the limits of our technique,
138
482761
3622
dok nismo zaista shvatili
granice naše tehnike,
granice naše tehnike,
zato što kada smo jednom shvatili granice,
08:18
because once we understood those limits,
139
486383
1950
08:20
we could figure out how to push them.
140
488333
2346
smislili bismo kako da ih gurnemo dalje.
08:22
And that led to experiments like this one,
141
490679
3181
A to nas je dovelo
do eksperimenta poput ovog,
do eksperimenta poput ovog,
08:25
where again, I'm going to speak
to a bag of chips,
to a bag of chips,
142
493860
2739
gde ja, ponovo, govorim kesi čipsa,
08:28
but this time we've moved our camera
about 15 feet away,
about 15 feet away,
143
496599
4830
ali ovog puta smo pomerili kameru
oko 4,5 m dalje,
oko 4,5 m dalje,
08:33
outside, behind a soundproof window,
144
501429
2833
van, iza zvučno izolovanog prozora,
08:36
and the whole thing is lit
by only natural sunlight.
by only natural sunlight.
145
504262
2803
i cela ova stvar je osvetljena samo
prirodnom svetlošću.
prirodnom svetlošću.
08:40
And so here's the video that we captured.
146
508529
2155
I evo snimka koji smo snimili.
08:44
And this is what things sounded like
from inside, next to the bag of chips.
from inside, next to the bag of chips.
147
512450
4559
A ovako se stvari čuju iznutra,
pored kese čipsa.
pored kese čipsa.
08:49
(Audio) Mary had a little lamb
whose fleece was white as snow,
whose fleece was white as snow,
148
517009
5038
(Audio) Meri je imala malo jagnje,
čije je runo bilo belo kao sneg
čije je runo bilo belo kao sneg
08:54
and everywhere that Mary went,
that lamb was sure to go.
that lamb was sure to go.
149
522047
5619
i kuda god je Meri išla,
jagnje je išlo s njom.
jagnje je išlo s njom.
08:59
AD: And here's what we were able
to recover from our silent video
to recover from our silent video
150
527666
4017
ED: A ovo smo uspeli da povratimo
sa našeg tihog snimka
sa našeg tihog snimka
09:03
captured outside behind that window.
151
531683
2345
snimljenog van, iza tog prozora.
09:06
(Audio) Mary had a little lamb
whose fleece was white as snow,
whose fleece was white as snow,
152
534028
4435
(Audio) Meri je imala malo jagnje
čije je runo bilo belo kao sneg,
čije je runo bilo belo kao sneg,
09:10
and everywhere that Mary went,
that lamb was sure to go.
that lamb was sure to go.
153
538463
5457
i kuda god je Meri išla,
jagnje je išlo s njom.
jagnje je išlo s njom.
09:15
(Applause)
154
543920
6501
(Aplauz)
09:22
AD: And there are other ways
that we can push these limits as well.
that we can push these limits as well.
155
550421
3542
ED: A postoje, takođe, i drugi načini
kako da pomerimo granice.
kako da pomerimo granice.
09:25
So here's a quieter experiment
156
553963
1798
Time, ovde imamo tiši eksperiment,
09:27
where we filmed some earphones
plugged into a laptop computer,
plugged into a laptop computer,
157
555761
4110
gde smo snimili jedne slušalice
priključene na laptop,
priključene na laptop,
09:31
and in this case, our goal was to recover
the music that was playing on that laptop
the music that was playing on that laptop
158
559871
4110
gde je naš cilj bio da muziku
koja je puštena na laptopu
koja je puštena na laptopu
09:35
from just silent video
159
563981
2299
povratimo sa tihog snimka
09:38
of these two little plastic earphones,
160
566280
2507
sa ove dve male plastične bubice,
09:40
and we were able to do this so well
161
568787
2183
i uspeli smo da uradimo to tako dobro
09:42
that I could even Shazam our results.
162
570970
2461
da smo mogli naše rezultate
koristiti čak i na Šazemu.
koristiti čak i na Šazemu.
09:45
(Laughter)
163
573431
2411
(Smeh)
09:49
(Music: "Under Pressure" by Queen)
164
577191
10034
(Muzika: "Pod pritiskom", grupa Kvin)
10:01
(Applause)
165
589615
4969
(Aplauz)
10:06
And we can also push things
by changing the hardware that we use.
by changing the hardware that we use.
166
594584
4551
A možemo i poboljšati stvari menjajući
hardver koji koristimo.
hardver koji koristimo.
10:11
Because the experiments
I've shown you so far
I've shown you so far
167
599135
2461
Jer su eksperimenti koje sam vam
pokazao do sada
pokazao do sada
10:13
were done with a camera,
a high-speed camera,
a high-speed camera,
168
601596
2322
napravljeni kamerom,
kamerom velike brzine,
kamerom velike brzine,
10:15
that can record video
about a 100 times faster
about a 100 times faster
169
603918
2879
koja može da napravi snimak
oko 100 puta brže
oko 100 puta brže
10:18
than most cell phones,
170
606797
1927
u odnosu na većinu mobilnih telefona,
10:20
but we've also found a way
to use this technique
to use this technique
171
608724
2809
ali takođe smo našli način da
upotrebimo ovu tehniku
upotrebimo ovu tehniku
10:23
with more regular cameras,
172
611533
2230
sa mnogo običnijim kamerama,
10:25
and we do that by taking advantage
of what's called a rolling shutter.
of what's called a rolling shutter.
173
613763
4069
i uradili smo to tako što smo iskoristili
nešto što se zove pokretni okidač.
nešto što se zove pokretni okidač.
10:29
You see, most cameras
record images one row at a time,
record images one row at a time,
174
617832
4798
Vidite, većina fotoaparata snima slike
jedan po jedan red,
jedan po jedan red,
10:34
and so if an object moves
during the recording of a single image,
during the recording of a single image,
175
622630
5702
tako da ako se predmet pomeri
tokom snimanja jedne slike,
tokom snimanja jedne slike,
10:40
there's a slight time delay
between each row,
between each row,
176
628344
2717
postoji kratko kašnjenje
između svakog reda
između svakog reda
10:43
and this causes slight artifacts
177
631061
3157
i zbog ovoga ostaju blagi tragovi
10:46
that get coded into each frame of a video.
178
634218
3483
koji se kodiraju u svaki frejm snimka.
10:49
And so what we found
is that by analyzing these artifacts,
is that by analyzing these artifacts,
179
637701
3806
Otkrili smo da analizom ovih tragova
10:53
we can actually recover sound
using a modified version of our algorithm.
using a modified version of our algorithm.
180
641507
4615
zapravo možemo povratiti zvuk koristeći
izmenjenu verziju našeg algoritma.
izmenjenu verziju našeg algoritma.
10:58
So here's an experiment we did
181
646122
1912
Evo eksperimenta koji smo uradili,
11:00
where we filmed a bag of candy
182
648034
1695
gde smo snimili kesicu bombona
11:01
while a nearby loudspeaker played
183
649729
1741
dok se iz zvučnika u blizini čulo
11:03
the same "Mary Had a Little Lamb"
music from before,
music from before,
184
651470
2972
ista muzika od pre:
"Meri je imala malo jagnje",
"Meri je imala malo jagnje",
11:06
but this time, we used just a regular
store-bought camera,
store-bought camera,
185
654442
4203
ali ovog puta smo koristili
običan fotoaparat iz prodavnice
običan fotoaparat iz prodavnice
11:10
and so in a second, I'll play for you
the sound that we recovered,
the sound that we recovered,
186
658645
3174
i sada ću vam pustiti
zvuk koji smo povratili,
zvuk koji smo povratili,
11:13
and it's going to sound
distorted this time,
distorted this time,
187
661819
2050
i ovog puta će zvučati izmenjeno,
11:15
but listen and see if you can still
recognize the music.
recognize the music.
188
663869
2836
ali slušajte i vidite da li još uvek
možete da prepoznate muziku.
možete da prepoznate muziku.
11:19
(Audio: "Mary Had a Little Lamb")
189
667723
6223
(Audio: "Meri je imala malo jagnje")
11:37
And so, again, that sounds distorted,
190
685527
3465
Opet, to zvuči izmenjeno,
11:40
but what's really amazing here
is that we were able to do this
is that we were able to do this
191
688992
4386
ali ovde je zaista neverovatno to
što smo ovo mogli da uradimo
što smo ovo mogli da uradimo
11:45
with something
that you could literally run out
that you could literally run out
192
693378
2626
sa nečim što bukvalno možete otići
11:48
and pick up at a Best Buy.
193
696004
1444
i kupiti u lokalnoj prodavnici
bele tehnike.
bele tehnike.
11:51
So at this point,
194
699122
1363
U ovom trenutku,
11:52
a lot of people see this work,
195
700485
1974
mnogi ljudi vide kako ovo radi
11:54
and they immediately think
about surveillance.
about surveillance.
196
702459
3413
i odmah pomisle na nadgledanje.
11:57
And to be fair,
197
705872
2415
Da budem iskren, nije teško zamisliti
12:00
it's not hard to imagine how you might use
this technology to spy on someone.
this technology to spy on someone.
198
708287
4133
kako biste mogli koristiti ovu tehnologiju
da špijunirate nekoga.
da špijunirate nekoga.
12:04
But keep in mind that there's already
a lot of very mature technology
a lot of very mature technology
199
712420
3947
Ali imajte na umu da trenutno već postoji
dosta veoma razvijene tehnologije
dosta veoma razvijene tehnologije
12:08
out there for surveillance.
200
716367
1579
za nadgledanje.
12:09
In fact, people have been using lasers
201
717946
2090
Zapravo, ljudi su decenijama
koristili lasere
koristili lasere
12:12
to eavesdrop on objects
from a distance for decades.
from a distance for decades.
202
720036
2799
kako bi prisluškivali sa udaljenosti.
12:15
But what's really new here,
203
723978
2025
Ali ono što je ovde zaista novo,
12:18
what's really different,
204
726003
1440
zaista drugačije,
12:19
is that now we have a way
to picture the vibrations of an object,
to picture the vibrations of an object,
205
727443
4295
je da sada imamo način da prikažemo
vibracije na predmetu,
vibracije na predmetu,
12:23
which gives us a new lens
through which to look at the world,
through which to look at the world,
206
731738
3413
što nam daje novi objektiv
kroz koji možemo gledati svet
kroz koji možemo gledati svet
12:27
and we can use that lens
207
735151
1510
i taj objektiv možemo koristiti
12:28
to learn not just about forces like sound
that cause an object to vibrate,
that cause an object to vibrate,
208
736661
4899
da saznamo, ne samo o silama poput zvuka
od kojih predmet vibrira,
od kojih predmet vibrira,
12:33
but also about the object itself.
209
741560
2288
nego i o samom predmetu.
12:36
And so I want to take a step back
210
744975
1693
Želim da se udaljim
12:38
and think about how that might change
the ways that we use video,
the ways that we use video,
211
746668
4249
i razmislim o tome kako to može
da promeni način na koji koristimo video,
da promeni način na koji koristimo video,
12:42
because we usually use video
to look at things,
to look at things,
212
750917
3553
jer video obično koristimo
da posmatramo stvari
da posmatramo stvari
12:46
and I've just shown you how we can use it
213
754470
2322
i pokazao sam vam upravo
kako ga možemo koristiti
kako ga možemo koristiti
12:48
to listen to things.
214
756792
1857
da slušamo stvari.
12:50
But there's another important way
that we learn about the world:
that we learn about the world:
215
758649
3971
Ali tu je važan način
na koji saznajemo stvari o svetu -
na koji saznajemo stvari o svetu -
12:54
that's by interacting with it.
216
762620
2275
kroz interakciju.
12:56
We push and pull and poke and prod things.
217
764895
3111
Guramo i vučemo
i pritiskamo i čačkamo stvari.
i pritiskamo i čačkamo stvari.
13:00
We shake things and see what happens.
218
768006
3181
Tresemo ih da vidimo šta će se desiti.
13:03
And that's something that video
still won't let us do,
still won't let us do,
219
771187
4273
A video nam još uvek ne dozvoljava to,
13:07
at least not traditionally.
220
775460
2136
makar ne u tradicionalnom smislu.
13:09
So I want to show you some new work,
221
777596
1950
Želim da vam pokažem neke nove radove,
13:11
and this is based on an idea I had
just a few months ago,
just a few months ago,
222
779546
2667
a ovo je zasnovano na zamisli
koju sam imao pre nekoliko meseci,
koju sam imao pre nekoliko meseci,
13:14
so this is actually the first time
I've shown it to a public audience.
I've shown it to a public audience.
223
782213
3301
ovo je zapravo prvi put
da to pokazujem pred publikom.
da to pokazujem pred publikom.
13:17
And the basic idea is that we're going
to use the vibrations in a video
to use the vibrations in a video
224
785514
5363
Osnovna ideja je da ćemo koristiti
vibracije u video snimku
vibracije u video snimku
13:22
to capture objects in a way
that will let us interact with them
that will let us interact with them
225
790877
4481
da snimimo predmete na način
koji će nam dozvoliti interkaciju sa njima
koji će nam dozvoliti interkaciju sa njima
13:27
and see how they react to us.
226
795358
1974
i da vidimo kako reaguju na nas.
13:31
So here's an object,
227
799120
1764
Evo predmeta,
13:32
and in this case, it's a wire figure
in the shape of a human,
in the shape of a human,
228
800884
3832
i u ovom slučaju to je žičana figura
u obliku čoveka,
u obliku čoveka,
13:36
and we're going to film that object
with just a regular camera.
with just a regular camera.
229
804716
3088
i taj predmet ćemo snimiti
običnim fotoaparatom.
običnim fotoaparatom.
Dakle, nema ničeg posebnog
u vezi sa ovim fotoaparatom.
u vezi sa ovim fotoaparatom.
13:39
So there's nothing special
about this camera.
about this camera.
230
807804
2124
13:41
In fact, I've actually done this
with my cell phone before.
with my cell phone before.
231
809928
2961
Zapravo, ovo sam pre radio
sa svojim telefonom.
sa svojim telefonom.
13:44
But we do want to see the object vibrate,
232
812889
2252
Ali želimo da vidimo da predmet vibrira,
13:47
so to make that happen,
233
815141
1133
i kako bi se to desilo,
13:48
we're just going to bang a little bit
on the surface where it's resting
on the surface where it's resting
234
816274
3346
samo malo ćemo udariti površinu
gde predmet stoji
gde predmet stoji
13:51
while we record this video.
235
819620
2138
dok snimamo ovaj video.
13:59
So that's it: just five seconds
of regular video,
of regular video,
236
827398
3671
To je to: samo pet sekundi
običnog video snimka,
običnog video snimka,
14:03
while we bang on this surface,
237
831069
2136
dok lupamo po površini
14:05
and we're going to use
the vibrations in that video
the vibrations in that video
238
833205
3513
i iskoristićemo vibracije iz tog videa,
14:08
to learn about the structural
and material properties of our object,
and material properties of our object,
239
836718
4544
kako bismo saznali strukturna
i materijalna svojstva našeg predmeta
i materijalna svojstva našeg predmeta
14:13
and we're going to use that information
to create something new and interactive.
to create something new and interactive.
240
841262
4834
i iskoristićemo te informacije
da stvorimo nešto novo i interaktivno.
da stvorimo nešto novo i interaktivno.
14:24
And so here's what we've created.
241
852866
2653
Evo šta smo stvorili.
14:27
And it looks like a regular image,
242
855519
2229
Izgleda kao obična slika,
14:29
but this isn't an image,
and it's not a video,
and it's not a video,
243
857748
3111
ali ovo nije slika i nije video,
14:32
because now I can take my mouse
244
860859
2368
jer sada mogu da uzmem miš
14:35
and I can start interacting
with the object.
with the object.
245
863227
2859
i da budem u interakciji
sa ovim predmetom.
sa ovim predmetom.
14:44
And so what you see here
246
872936
2357
Ovde možete videti
14:47
is a simulation of how this object
247
875389
2226
simulaciju toga kako bi ovaj predmet
14:49
would respond to new forces
that we've never seen before,
that we've never seen before,
248
877615
4458
odgovorio na nove sile
koje do sada nismo videli,
koje do sada nismo videli,
14:54
and we created it from just
five seconds of regular video.
five seconds of regular video.
249
882073
3633
i to smo stvorili iz samo
pet sekundi običnog video snimka.
pet sekundi običnog video snimka.
14:59
(Applause)
250
887249
4715
(Aplauz)
15:09
And so this is a really powerful
way to look at the world,
way to look at the world,
251
897421
3227
Ovo je zaista moćan način
da se posmatra svet
da se posmatra svet
15:12
because it lets us predict
how objects will respond
how objects will respond
252
900648
2972
jer nam dozvoljava da predvidimo
kako će predmeti odgovoriti
kako će predmeti odgovoriti
15:15
to new situations,
253
903620
1823
na nove situacije
15:17
and you could imagine, for instance,
looking at an old bridge
looking at an old bridge
254
905443
3473
i možete da zamislite, na primer,
kako gledate stari most
kako gledate stari most
15:20
and wondering what would happen,
how would that bridge hold up
how would that bridge hold up
255
908916
3527
i pitate se šta bi se desilo,
kako bi se ponašao taj most
kako bi se ponašao taj most
15:24
if I were to drive my car across it.
256
912443
2833
kada bih prešao preko njega
svojim automobilom.
svojim automobilom.
15:27
And that's a question
that you probably want to answer
that you probably want to answer
257
915276
2774
A to je pitanje na koje
verovatno želite da odgovorite
verovatno želite da odgovorite
15:30
before you start driving
across that bridge.
across that bridge.
258
918050
2560
pre nego što počnete da vozite
preko tog mosta.
preko tog mosta.
15:33
And of course, there are going to be
limitations to this technique,
limitations to this technique,
259
921988
3272
Naravno, biće nekoliko ograničenja
za ovu tehniku,
za ovu tehniku,
15:37
just like there were
with the visual microphone,
with the visual microphone,
260
925260
2462
kao što ih je bilo
i sa vizuelnim mikrofonom,
i sa vizuelnim mikrofonom,
15:39
but we found that it works
in a lot of situations
in a lot of situations
261
927722
3181
ali otkrili smo da funkcioniše
u mnogim situacijama
u mnogim situacijama
15:42
that you might not expect,
262
930903
1875
gde to ne biste očekivali,
15:44
especially if you give it longer videos.
263
932778
2768
naročito ako radite sa dužim snimcima,
15:47
So for example,
here's a video that I captured
here's a video that I captured
264
935546
2508
Na primer,
evo video zapisa koji sam napravio
evo video zapisa koji sam napravio
15:50
of a bush outside of my apartment,
265
938054
2299
gde je žbun ispred mog stana,
15:52
and I didn't do anything to this bush,
266
940353
3088
a njemu nisam radio ništa
15:55
but by capturing a minute-long video,
267
943441
2705
osim što sam ga snimao jedan minut,
15:58
a gentle breeze caused enough vibrations
268
946146
3378
lagan povetarac
izazvao je dovoljno vibracija
izazvao je dovoljno vibracija
16:01
that we could learn enough about this bush
to create this simulation.
to create this simulation.
269
949524
3587
da smo mogli da saznamo dovoljno
o ovom žbunu da napravimo simulaciju.
o ovom žbunu da napravimo simulaciju.
16:07
(Applause)
270
955270
6142
(Aplauz)
16:13
And so you could imagine giving this
to a film director,
to a film director,
271
961412
2972
Možete da zamislite da ovo date
filmskom režiseru
filmskom režiseru
16:16
and letting him control, say,
272
964384
1719
i date mu da kontroliše, na primer,
16:18
the strength and direction of wind
in a shot after it's been recorded.
in a shot after it's been recorded.
273
966103
4922
jačinu i pravac vetra u snimku
nakon što je on nastao.
nakon što je on nastao.
16:24
Or, in this case, we pointed our camera
at a hanging curtain,
at a hanging curtain,
274
972810
4535
Ili, u ovom slučaju, uperili smo
fotoaparat u zavesu koja visi
fotoaparat u zavesu koja visi
16:29
and you can't even see
any motion in this video,
any motion in this video,
275
977345
4129
i u ovom snimku čak ni ne možete
da vidite bilo kakvo kretanje,
da vidite bilo kakvo kretanje,
16:33
but by recording a two-minute-long video,
276
981474
2925
ali snimanjem videa dugog dva minuta,
16:36
natural air currents in this room
277
984399
2438
prirodne vazdušne struje u prostoriji
16:38
created enough subtle,
imperceptible motions and vibrations
imperceptible motions and vibrations
278
986837
4412
stvorile su dovoljno suptilnog,
neprimetnog kretanja i vibracija
neprimetnog kretanja i vibracija
16:43
that we could learn enough
to create this simulation.
to create this simulation.
279
991249
2565
da možemo da saznamo dovoljno
da stvorimo ovu simulaciju.
da stvorimo ovu simulaciju.
16:48
And ironically,
280
996243
2366
Ironično je to
16:50
we're kind of used to having
this kind of interactivity
this kind of interactivity
281
998609
3088
kako smo nekako priviknuti da imamo
ovakvu vrstu interaktivnosti
ovakvu vrstu interaktivnosti
16:53
when it comes to virtual objects,
282
1001697
2647
kod virtuelnih predmeta,
16:56
when it comes to video games
and 3D models,
and 3D models,
283
1004344
3297
što se tiče video igara i 3D modela,
16:59
but to be able to capture this information
from real objects in the real world
from real objects in the real world
284
1007641
4404
ali kako bismo dobili ove informacije
iz pravih predmeta iz stvarnosti,
iz pravih predmeta iz stvarnosti,
17:04
using just simple, regular video,
285
1012045
2817
koristeći proste, obične video snimke,
17:06
is something new that has
a lot of potential.
a lot of potential.
286
1014862
2183
to je nešto novo
što ima dosta potencijala.
što ima dosta potencijala.
17:10
So here are the amazing people
who worked with me on these projects.
who worked with me on these projects.
287
1018410
4904
Evo neverovatnih ljudi koji su sa mnom
radili na ovom projektu.
radili na ovom projektu.
17:16
(Applause)
288
1024057
5596
(Aplauz)
17:24
And what I've shown you today
is only the beginning.
is only the beginning.
289
1032819
3057
Ono što sam vam pokazao danas
je samo početak.
je samo početak.
17:27
We've just started to scratch the surface
290
1035876
2113
Samo smo zagrebali površinu toga
17:29
of what you can do
with this kind of imaging,
with this kind of imaging,
291
1037989
2972
što možete da uradite
sa ovim načinom prikazivanja slika,
sa ovim načinom prikazivanja slika,
17:32
because it gives us a new way
292
1040961
2286
jer nam to daje novi metod
17:35
to capture our surroundings
with common, accessible technology.
with common, accessible technology.
293
1043342
4724
da snimimo svoju okolinu,
prostom, dostupnom tehnologijom.
prostom, dostupnom tehnologijom.
17:40
And so looking to the future,
294
1048066
1929
Gledajući u budućnost,
17:41
it's going to be
really exciting to explore
really exciting to explore
295
1049995
2037
biće zaista zanimljiva
za istraživanje toga
za istraživanje toga
17:44
what this can tell us about the world.
296
1052032
1856
šta nam ovo može reći o svetu.
17:46
Thank you.
297
1054381
1204
Hvala vam.
17:47
(Applause)
298
1055610
6107
(Aplauz)
ABOUT THE SPEAKER
Abe Davis - Computer scientistComputer vision expert Abe Davis pioneers methods to extract audio from silent digital videos, even footage shot on ordinary consumer cameras.
Why you should listen
MIT PhD student, computer vision wizard and rap artist Abe Davis has co-created the world’s most improbable audio instrument. In 2014, Davis and his collaborators debuted the “visual microphone,” an algorithm that samples the sympathetic vibrations of ordinary objects (such as a potato chip bag) from ordinary high-speed video footage and transduces them into intelligible audio tracks.
Davis is also the author of Caperture, a 3D-imaging app designed to create and share 3D images on any compatible smartphone.
Abe Davis | Speaker | TED.com