TED2007
Blaise Agüera y Arcas: How PhotoSynth can connect the world's images
Blaise Aguera y Arcas demonstrira Photosynth
Filmed:
Readability: 4.3
5,831,957 views
Blaisse Aguera y Arcas vodi sjajnu demonstraciju Photosynth-a, softvera koji bi mogao promijeniti način na koji promatramo digitalne slike. Koristeći fotografije odabrane s Weba, Photosynth gradi pejzaže iz sna koji oduzimaju dah i dopušta nam da se krećemo kroz njih.
Blaise Agüera y Arcas - Software architect
Blaise Agüera y Arcas works on machine learning at Google. Previously a Distinguished Engineer at Microsoft, he has worked on augmented reality, mapping, wearable computing and natural user interfaces. Full bio
Blaise Agüera y Arcas works on machine learning at Google. Previously a Distinguished Engineer at Microsoft, he has worked on augmented reality, mapping, wearable computing and natural user interfaces. Full bio
Double-click the English transcript below to play the video.
00:25
What I'm going to show you first, as quickly as I can,
0
0
2000
Ono što ću vam prvo pokazati, što brže mogu,
00:27
is some foundational work, some new technology
1
2000
4000
jest neki osnovni rad, neka nova tehnologija
00:31
that we brought to Microsoft as part of an acquisition
2
6000
3000
koju smo donijeli u Microsoft kao dio akvizicije
00:34
almost exactly a year ago. This is Seadragon,
3
9000
3000
skoro točno prije godinu dana. Ovo je Seadragon.
00:37
and it's an environment in which you can either locally or remotely
4
12000
3000
I to je okoliš u kojem možete lokalno ili udaljeno
00:40
interact with vast amounts of visual data.
5
15000
3000
biti u interakciji s ogromnom količinom vizualnih podataka.
00:43
We're looking at many, many gigabytes of digital photos here
6
18000
3000
Gledamo mnoštvo i mnoštvo gigabajta digitalnih fotografija
00:46
and kind of seamlessly and continuously zooming in,
7
21000
3000
i nekako neprimjetno i kontinuirano povećavamo,
00:50
panning through the thing, rearranging it in any way we want.
8
25000
2000
pomičemo se, organiziramo ih kako god želimo.
00:52
And it doesn't matter how much information we're looking at,
9
27000
4000
I nije važno koliku količinu informacija promatramo,
00:56
how big these collections are or how big the images are.
10
31000
3000
nije važno kolike su te kolekcije ili kolike su slike.
00:59
Most of them are ordinary digital camera photos,
11
34000
2000
Većina njih su obične fotografije s digitalne kamere,
01:01
but this one, for example, is a scan from the Library of Congress,
12
36000
3000
ali ova na primjer, je sken iz knjižnice Kongresa,
01:05
and it's in the 300 megapixel range.
13
40000
2000
i nalazi se u rasponu od 300 megapiksela.
01:08
It doesn't make any difference
14
43000
1000
Nema nikakve razlike
01:09
because the only thing that ought to limit the performance
15
44000
3000
jer jedina stvar koja bi trebala ograničavati performanse
01:12
of a system like this one is the number of pixels on your screen
16
47000
3000
sustava poput ovog je broj piksela na vašem ekranu
01:15
at any given moment. It's also very flexible architecture.
17
50000
3000
u bilo kojem datom trenutku. To je također vrlo fleksibilna arhitektura.
01:18
This is an entire book, so this is an example of non-image data.
18
53000
3000
Ovo je čitava knjiga, primjer podataka koji nisu slike.
01:22
This is "Bleak House" by Dickens. Every column is a chapter.
19
57000
5000
Ovo je Hladna kuća od Dickensa. Svaki stupac je poglavlje.
01:27
To prove to you that it's really text, and not an image,
20
62000
4000
Kako bih vam dokazao da je to stvarno tekst, a ne slika,
01:31
we can do something like so, to really show
21
66000
2000
možemo učiniti nešto poput ovoga, da zaista pokažemo
01:33
that this is a real representation of the text; it's not a picture.
22
68000
3000
da je ovo stvarni prikaz teksta, da nije slika.
01:37
Maybe this is a kind of an artificial way to read an e-book.
23
72000
2000
Možda je ovo pomalo umjetan način da se čita e-knjiga.
01:39
I wouldn't recommend it.
24
74000
1000
Ne bih ga preporučio.
01:40
This is a more realistic case. This is an issue of The Guardian.
25
75000
3000
Ovo je malo realističniji slučaj. Ovo je primjerak Guardian-a.
01:43
Every large image is the beginning of a section.
26
78000
2000
Svaka velika slika je početak odjeljka.
01:45
And this really gives you the joy and the good experience
27
80000
3000
I ovo zaista daje radost i dobro iskustvo
01:48
of reading the real paper version of a magazine or a newspaper,
28
83000
5000
čitanja prave papirnate verzije časopisa ili novina,
01:54
which is an inherently multi-scale kind of medium.
29
89000
1000
koji su sami po sebi medij s više razina.
01:56
We've also done a little something
30
91000
1000
Također smo napravili ponešto
01:57
with the corner of this particular issue of The Guardian.
31
92000
3000
s uglom ovog primjerka Guardian-a.
02:00
We've made up a fake ad that's very high resolution --
32
95000
3000
Osmislili smo lažni oglas koji je vrlo visoke rezolucije --
02:03
much higher than you'd be able to get in an ordinary ad --
33
98000
2000
puno više nego u običnom oglasu --
02:05
and we've embedded extra content.
34
100000
2000
i ugradili smo dodatne sadržaje.
02:07
If you want to see the features of this car, you can see it here.
35
102000
2000
Ako želite vidjeti osobine ovog auta, možete ih vidjeti ovdje.
02:10
Or other models, or even technical specifications.
36
105000
4000
Ili druge modele, ili čak tehničke specifikacije.
02:15
And this really gets at some of these ideas
37
110000
2000
I ovo stvarno pokreće neke od tih ideja
02:18
about really doing away with those limits on screen real estate.
38
113000
4000
da se uklone granice oglašavanja nekretnina na ekranima.
02:22
We hope that this means no more pop-ups
39
117000
2000
Nadamo se da će to značiti kraj prozora koji iskaču
02:24
and other kind of rubbish like that -- shouldn't be necessary.
40
119000
2000
i druge vrste smeća poput tog -- to ne bi bilo potrebno.
02:27
Of course, mapping is one of those really obvious applications
41
122000
2000
Naravno, kartografija je jedna od zaista očitih primjena
02:29
for a technology like this.
42
124000
2000
za ovakvu tehnologiju.
02:31
And this one I really won't spend any time on,
43
126000
2000
Na ovo uopće neću trošiti vrijeme,
02:33
except to say that we have things to contribute to this field as well.
44
128000
2000
samo ću reći da imamo što doprinjeti i na ovom polju.
02:37
But those are all the roads in the U.S.
45
132000
2000
Ali ovo su sve ceste u Americi
02:39
superimposed on top of a NASA geospatial image.
46
134000
4000
postavljene iznad NASA-ine geoprostorne slike.
02:44
So let's pull up, now, something else.
47
139000
2000
Podignimo, sada nešto drugo.
02:46
This is actually live on the Web now; you can go check it out.
48
141000
3000
Ovo je zapravo živo na internetu sada, možete ga otići provjeriti.
02:49
This is a project called Photosynth,
49
144000
1000
Ovo je projekt koji se zove Photosynth,
02:51
which really marries two different technologies.
50
146000
1000
koji združuje dvije različite tehnologije.
02:52
One of them is Seadragon
51
147000
1000
Jedna od njih je Seadragon
02:54
and the other is some very beautiful computer vision research
52
149000
2000
a druga je prekrasno istraživanje vezano uz računalni vid
02:57
done by Noah Snavely, a graduate student at the University of Washington,
53
152000
2000
kojim se bavi Noah Snavely, apsolvent na Sveučilištu u Washingtonu,
03:00
co-advised by Steve Seitz at U.W.
54
155000
2000
pod savjetovanjem od strane Steve Seitza na Sveučilištu u Washingtonu
03:02
and Rick Szeliski at Microsoft Research. A very nice collaboration.
55
157000
4000
i Ricka Szeilski pri Microsoft istraživanjima. Vrlo dobra suradnja.
03:07
And so this is live on the Web. It's powered by Seadragon.
56
162000
2000
Dakle ovo je uživo na Internetu. Pokreće ga Seadragon.
03:09
You can see that when we kind of do these sorts of views,
57
164000
2000
Možete vidjeti to kad radimo ovakve vrste pogleda,
03:12
where we can dive through images
58
167000
1000
gdje možemo zaroniti kroz slike
03:14
and have this kind of multi-resolution experience.
59
169000
1000
i imamo ovakvo više-rezolucijsko iskustvo.
03:16
But the spatial arrangement of the images here is actually meaningful.
60
171000
4000
Ali prostorni raspored slika ovdje je zaista smislen.
03:20
The computer vision algorithms have registered these images together
61
175000
3000
Algoritmi za računalni vid su registrirali ove slike skupa,
03:23
so that they correspond to the real space in which these shots --
62
178000
4000
tako da odgovaraju stvarnom prostoru u kojem su ove slike --
03:27
all taken near Grassi Lakes in the Canadian Rockies --
63
182000
2000
sve slikane blizu jezera Grassi u kanadskom Stjenjaku --
03:31
all these shots were taken. So you see elements here
64
186000
2000
napravljene. Dakle ovdje vidite elemente
03:33
of stabilized slide-show or panoramic imaging,
65
188000
4000
stabilizirane prezentacije ili panoramskog slikanja,
03:40
and these things have all been related spatially.
66
195000
2000
i ovo su sve stvari koje su prostorno povezane.
03:42
I'm not sure if I have time to show you any other environments.
67
197000
3000
Nisam siguran imam li vremena da vam pokažem neke druge okoliše.
03:45
There are some that are much more spatial.
68
200000
1000
Koji su puno prostraniji.
03:47
I would like to jump straight to one of Noah's original data-sets --
69
202000
3000
Želio bih odmah prijeći na jedne od Noinih originalnih setova podataka --
03:50
and this is from an early prototype of Photosynth
70
205000
2000
i ovo je iz ranog prototipa Photosyntha
03:52
that we first got working in the summer --
71
207000
2000
koji smo prvo počeli raditi u ljeto --
03:54
to show you what I think
72
209000
1000
da vam pokažem što mislim
03:55
is really the punch line behind this technology,
73
210000
3000
što je zapravo ponta ove tehnologije,
03:59
the Photosynth technology. And it's not necessarily so apparent
74
214000
2000
Photosynth tehnologije. I ona nije nužno tako očita
04:01
from looking at the environments that we've put up on the website.
75
216000
3000
kao gledanje okoliša koje smo stavili na web stranicu.
04:04
We had to worry about the lawyers and so on.
76
219000
2000
Morali smo se brinuti za odvjetnike i tako dalje.
04:07
This is a reconstruction of Notre Dame Cathedral
77
222000
1000
Ovo je rekonstrukcija katedrale Notre Dame
04:09
that was done entirely computationally
78
224000
2000
koja je napravljena potpuno računalno
04:11
from images scraped from Flickr. You just type Notre Dame into Flickr,
79
226000
3000
od slika skupljenih na Flickr-u. Samo utipkate Notre Dame u Flickr,
04:14
and you get some pictures of guys in t-shirts, and of the campus
80
229000
3000
i dobijete slike ljudi u majicama, i sveučilišta
04:17
and so on. And each of these orange cones represents an image
81
232000
4000
i tako dalje. I svaki od ovih narančastih čunjeva predstavlja sliku
04:22
that was discovered to belong to this model.
82
237000
2000
za koju se otkrilo da pripada ovom modelu.
04:26
And so these are all Flickr images,
83
241000
2000
Tako da su ovo sve slike s Flickr-a,
04:28
and they've all been related spatially in this way.
84
243000
3000
i one su sve povezane prostorno ovim putem.
04:31
And we can just navigate in this very simple way.
85
246000
2000
I možemo se kretati kroz njih na ovaj jednostavan način.
04:35
(Applause)
86
250000
9000
(Pljesak)
04:44
You know, I never thought that I'd end up working at Microsoft.
87
259000
2000
Znate, nikad nisam mislio da ću raditi u Microsoftu.
04:46
It's very gratifying to have this kind of reception here.
88
261000
4000
Zaista je lijepo biti primljen na ovaj način ovdje.
04:50
(Laughter)
89
265000
3000
(Smijeh)
04:53
I guess you can see
90
268000
3000
Pretpostavljam da možete vidjeti
04:56
this is lots of different types of cameras:
91
271000
2000
da je ovo mnoštvo različitih kamera:
04:58
it's everything from cell phone cameras to professional SLRs,
92
273000
3000
tu imamo sve od kamera s mobilnih telefona do profesionalnih fotoaparata,
05:02
quite a large number of them, stitched
93
277000
1000
zaista velik broj njih, povezanih
05:03
together in this environment.
94
278000
1000
u ovom okolišu.
05:04
And if I can, I'll find some of the sort of weird ones.
95
279000
2000
I ako uspijem, pronaći ću neke od zaista čudnih.
05:08
So many of them are occluded by faces, and so on.
96
283000
3000
Mnoge od njih su zaklonjene licima, i tako dalje.
05:13
Somewhere in here there are actually
97
288000
1000
Negdje ovdje imamo zapravo
05:15
a series of photographs -- here we go.
98
290000
1000
serije fotografija -- evo nas.
05:17
This is actually a poster of Notre Dame that registered correctly.
99
292000
3000
Ovo je zapravo plakat Notre Dame koji se registrirao ispravno.
05:21
We can dive in from the poster
100
296000
2000
Možemo zaroniti iz plakata
05:24
to a physical view of this environment.
101
299000
3000
u fizički pogled na ovaj okoliš.
05:31
What the point here really is is that we can do things
102
306000
3000
Ono što je zapravo poanta ovdje, je to da možemo činiti razne stvari
05:34
with the social environment. This is now taking data from everybody --
103
309000
5000
sa socijalnim okolišem. Ovdje uzimamo podatke od svih --
05:39
from the entire collective memory
104
314000
1000
od cjelokupne kolektivne memorije
05:40
of, visually, of what the Earth looks like --
105
315000
2000
onoga kako Zemlja izgleda --
05:43
and link all of that together.
106
318000
1000
i sve smo to povezali.
05:44
All of those photos become linked together,
107
319000
2000
Sve ove slike su postale povezane,
05:46
and they make something emergent
108
321000
1000
da bi stvorile nešto što izranja
05:47
that's greater than the sum of the parts.
109
322000
2000
što je veće od zbroja svih dijelova.
05:49
You have a model that emerges of the entire Earth.
110
324000
2000
Imate model koji izranja iz čitave Zemlje.
05:51
Think of this as the long tail to Stephen Lawler's Virtual Earth work.
111
326000
5000
Razmišljajte o ovome kao o nastavku rada Stephena Lawlera na Virtualnoj zemlji.
05:56
And this is something that grows in complexity
112
331000
2000
I ovo je nešto čemu složenost raste
05:58
as people use it, and whose benefits become greater
113
333000
3000
kako ga ljudi koriste, i čije povlastice postaju veće
06:01
to the users as they use it.
114
336000
2000
za korisnike kako ga koriste.
06:03
Their own photos are getting tagged with meta-data
115
338000
2000
Njihove slike se označavaju meta podatcima
06:05
that somebody else entered.
116
340000
1000
koje je netko drugi unio.
06:07
If somebody bothered to tag all of these saints
117
342000
3000
Ako se netko potrudio označiti sve ove svece
06:10
and say who they all are, then my photo of Notre Dame Cathedral
118
345000
3000
i reći tko su oni, onda je moja slika katedrale Notre Dame
06:13
suddenly gets enriched with all of that data,
119
348000
2000
iznenada postala obogaćena svim ovim podatcima,
06:15
and I can use it as an entry point to dive into that space,
120
350000
3000
i ja ju mogu koristiti kao točku ulaza kako bih zaronio u taj prostor,
06:18
into that meta-verse, using everybody else's photos,
121
353000
2000
u taj meta svemir, koristeći fotografije svih drugih,
06:21
and do a kind of a cross-modal
122
356000
2000
kako bih napravio isprepleteni model
06:25
and cross-user social experience that way.
123
360000
3000
i isprepleteno korisničko iskustvo na ovaj način.
06:28
And of course, a by-product of all of that
124
363000
1000
I naravno, kao nusproizvod svega toga
06:30
is immensely rich virtual models
125
365000
2000
strašno bogate virtualne modele
06:32
of every interesting part of the Earth, collected
126
367000
2000
svakog zanimljivog dijela Zemlje, prikupljenih
06:35
not just from overhead flights and from satellite images
127
370000
3000
ne samo slikama iz aviona ili satelitskim slikama
06:38
and so on, but from the collective memory.
128
373000
2000
i tako dalje, već i iz kolektivne memorije.
06:40
Thank you so much.
129
375000
2000
Hvala vam puno.
06:42
(Applause)
130
377000
11000
(Pljesak)
06:53
Chris Anderson: Do I understand this right? That what your software is going to allow,
131
388000
4000
Chris Anderson: Jesam li dobro shvatio? Da će ovo što tvoj softver nudi dozvoliti,
06:58
is that at some point, really within the next few years,
132
393000
2000
da u jednom trenutku, zapravo u sljedećih nekoliko godina,
07:01
all the pictures that are shared by anyone across the world
133
396000
4000
sve slike koje se dijele bilo gdje u svijetu
07:05
are going to basically link together?
134
400000
2000
budu povezane?
07:07
BAA: Yes. What this is really doing is discovering.
135
402000
2000
BAA: Da. Ono što ovo zapravo čini je otkriva.
07:09
It's creating hyperlinks, if you will, between images.
136
404000
3000
Kreira hiper veze, da tako kažem, između slika.
07:12
And it's doing that
137
407000
1000
I to čini
07:13
based on the content inside the images.
138
408000
1000
na osnovu sadražaja na slikama.
07:14
And that gets really exciting when you think about the richness
139
409000
3000
I to postaje zaista uzbudljivo kad razmišljaš o bogatstvu
07:17
of the semantic information that a lot of those images have.
140
412000
2000
jezikoslovnih informacija koje puno ovih slika ima.
07:19
Like when you do a web search for images,
141
414000
2000
Poput toga kada tražite slike na Internetu,
07:22
you type in phrases, and the text on the web page
142
417000
2000
samo upišete frazu, i tekst na web stranici
07:24
is carrying a lot of information about what that picture is of.
143
419000
3000
sadrži puno podataka o tome o čemu je slika.
07:27
Now, what if that picture links to all of your pictures?
144
422000
2000
A što ako se ta slika poveže sa svim vašim slikama?
07:29
Then the amount of semantic interconnection
145
424000
2000
Onda je količina jezikoslovne povezanosti
07:31
and the amount of richness that comes out of that
146
426000
1000
i količina bogatstva koja proizlazi iz toga
07:32
is really huge. It's a classic network effect.
147
427000
3000
zaista ogromna. To je klasičan efekt mreže.
07:35
CA: Blaise, that is truly incredible. Congratulations.
148
430000
2000
CA: Blaise, ovo je zaista nevjerojatno. Čestitam.
07:37
BAA: Thanks so much.
149
432000
1000
BAA: Hvala vam jako puno.
ABOUT THE SPEAKER
Blaise Agüera y Arcas - Software architectBlaise Agüera y Arcas works on machine learning at Google. Previously a Distinguished Engineer at Microsoft, he has worked on augmented reality, mapping, wearable computing and natural user interfaces.
Why you should listen
Blaise Agüera y Arcas is principal scientist at Google, where he leads a team working on machine intelligence for mobile devices. His group works extensively with deep neural nets for machine perception and distributed learning, and it also investigates so-called "connectomics" research, assessing maps of connections within the brain.
Agüera y Arcas' background is as multidimensional as the visions he helps create. In the 1990s, he authored patents on both video compression and 3D visualization techniques, and in 2001, he made an influential computational discovery that cast doubt on Gutenberg's role as the father of movable type.
He also created Seadragon (acquired by Microsoft in 2006), the visualization technology that gives Photosynth its amazingly smooth digital rendering and zoom capabilities. Photosynth itself is a vastly powerful piece of software capable of taking a wide variety of images, analyzing them for similarities, and grafting them together into an interactive three-dimensional space. This seamless patchwork of images can be viewed via multiple angles and magnifications, allowing us to look around corners or “fly” in for a (much) closer look. Simply put, it could utterly transform the way we experience digital images.
He joined Microsoft when Seadragon was acquired by Live Labs in 2006. Shortly after the acquisition of Seadragon, Agüera y Arcas directed his team in a collaboration with Microsoft Research and the University of Washington, leading to the first public previews of Photosynth several months later. His TED Talk on Seadragon and Photosynth in 2007 is rated one of TED's "most jaw-dropping." He returned to TED in 2010 to demo Bing’s augmented reality maps.
Fun fact: According to the author, Agüera y Arcas is the inspiration for the character Elgin in the 2012 best-selling novel Where'd You Go, Bernadette?
Blaise Agüera y Arcas | Speaker | TED.com