TED2007
Blaise Agüera y Arcas: How PhotoSynth can connect the world's images
Blaise Agura y Arcas face o demonstraţie în Photosynth
Filmed:
Readability: 4.3
5,831,957 views
Blaise Agura y Arcas face o demonstraţie uluitoare în Photosynth, un program care ne-ar putea transforma percepţia asupra imaginilor digitale. Folosind imagini statice culese de pe Web, Photosynth creează panorame care iţi taie respiraţia şi ne permite să navigăm prin ele.
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
Ceea ce vreau să vă arăt mai întâi, foarte rapid,
00:27
is some foundational work, some new technology
1
2000
4000
este o lucrare fundamentală, o nouă tehnologie
00:31
that we brought to Microsoft as part of an acquisition
2
6000
3000
pe care am adus-o la Microsoft ca parte dintr-o achiziţie
00:34
almost exactly a year ago. This is Seadragon,
3
9000
3000
aproximativ acum un an. Acesta este Seadragon.
00:37
and it's an environment in which you can either locally or remotely
4
12000
3000
Este un mediu care poate interacţiona fie local fie la distantă
00:40
interact with vast amounts of visual data.
5
15000
3000
cu o mare cantitate de date vizuale.
00:43
We're looking at many, many gigabytes of digital photos here
6
18000
3000
Ne uităm acum la mulţi, mulţi gigabaiţi de poze digitale
00:46
and kind of seamlessly and continuously zooming in,
7
21000
3000
si mărind în continuu aproape perfect,
00:50
panning through the thing, rearranging it in any way we want.
8
25000
2000
panoramând prin aceasta, rearanjând-o oricum dorim.
00:52
And it doesn't matter how much information we're looking at,
9
27000
4000
Nu conteaza la cât de multă informaţie ne uităm,
00:56
how big these collections are or how big the images are.
10
31000
3000
cât de mari sunt aceste colecţii sau cât de mari sunt imaginile.
00:59
Most of them are ordinary digital camera photos,
11
34000
2000
Majoritatea sunt poze făcute cu o camera digitală obişnuită,
01:01
but this one, for example, is a scan from the Library of Congress,
12
36000
3000
dar aceasta, de exemplu, este o scanare de la Biblioteca Congresului,
01:05
and it's in the 300 megapixel range.
13
40000
2000
de ordinul a 300 megapixeli.
01:08
It doesn't make any difference
14
43000
1000
Nu are importanţă
01:09
because the only thing that ought to limit the performance
15
44000
3000
pentru că singurul lucru care ar trebui să limiteze performanţa
01:12
of a system like this one is the number of pixels on your screen
16
47000
3000
unui asemenea sistem este numărul de pixeli al monitorului vostru
01:15
at any given moment. It's also very flexible architecture.
17
50000
3000
la un moment dat. Este de asemenea o arhitectură foarte flexibilă.
01:18
This is an entire book, so this is an example of non-image data.
18
53000
3000
Aceasta este o carte intreagă, un exemplu de dată non-imaginativă.
01:22
This is "Bleak House" by Dickens. Every column is a chapter.
19
57000
5000
Este vorba despre "Casa pustie" (Bleak House) a lui Charles Dickens. Fiecare coloană reprezintă un capitol.
01:27
To prove to you that it's really text, and not an image,
20
62000
4000
Pentru a vă demonstra că este vorba despre text, şi nu o imagine,
01:31
we can do something like so, to really show
21
66000
2000
putem face aşa ceva, pentru a arată cu adevarat
01:33
that this is a real representation of the text; it's not a picture.
22
68000
3000
că este o reprezentare reală a unui text; nu este o poză.
01:37
Maybe this is a kind of an artificial way to read an e-book.
23
72000
2000
Poate acesta este un mod artificial de a citi o carte în format electronic.
01:39
I wouldn't recommend it.
24
74000
1000
Nu v-aş recomanda.
01:40
This is a more realistic case. This is an issue of The Guardian.
25
75000
3000
Acesta este un caz mai real. Este o ediţie a "Gardianul"-ui.
01:43
Every large image is the beginning of a section.
26
78000
2000
Fiecare imagine reprezintă începutul unei secţiuni.
01:45
And this really gives you the joy and the good experience
27
80000
3000
Acest lucru iţi dă într-adevar bucuria unei experienţe plăcute
01:48
of reading the real paper version of a magazine or a newspaper,
28
83000
5000
de a citi versiunea tiparită a unei reviste sau a unui ziar,
01:54
which is an inherently multi-scale kind of medium.
29
89000
1000
lucru care este inerent un mediu multi-dimensional.
01:56
We've also done a little something
30
91000
1000
Am făcut şi noi ceva micuţ
01:57
with the corner of this particular issue of The Guardian.
31
92000
3000
cu colţul acestui număr din "Gardianul".
02:00
We've made up a fake ad that's very high resolution --
32
95000
3000
Am creat o reclamă falsă la o rezoluţie foarte mare --
02:03
much higher than you'd be able to get in an ordinary ad --
33
98000
2000
mult mai mare decât în orice reclamă obişnuită --
02:05
and we've embedded extra content.
34
100000
2000
şi am încorporat un conţinut suplimentar.
02:07
If you want to see the features of this car, you can see it here.
35
102000
2000
Dacă doriţi să vedeţi proprietăţile acestei maşini, le puteţi vedea aici.
02:10
Or other models, or even technical specifications.
36
105000
4000
Sau alte modele, sau chiar specificaţiile tehnice.
02:15
And this really gets at some of these ideas
37
110000
2000
Şi acest lucru chiar prinde la anumite idei,
02:18
about really doing away with those limits on screen real estate.
38
113000
4000
înlăturarea limitelor unui monitor.
02:22
We hope that this means no more pop-ups
39
117000
2000
Sperăm ca aceasta să însemne eliminarea pop-up-urilor
02:24
and other kind of rubbish like that -- shouldn't be necessary.
40
119000
2000
şi a altor mizerii de genul acesta -- care nu sunt necesare.
02:27
Of course, mapping is one of those really obvious applications
41
122000
2000
Bineînţeles, maparea este una dintre aplicaţiile evidente
02:29
for a technology like this.
42
124000
2000
pentru o tehnologie de genul acesta.
02:31
And this one I really won't spend any time on,
43
126000
2000
Nu voi zăbovi mult timp asupra acesteia,
02:33
except to say that we have things to contribute to this field as well.
44
128000
2000
ci doar ca să vă spun ca mai avem de contribuit la acest domeniu.
02:37
But those are all the roads in the U.S.
45
132000
2000
Acestea sunt toate drumurile din Statele Unite
02:39
superimposed on top of a NASA geospatial image.
46
134000
4000
suprapuse pe o imagine geospaţială de la NASA.
02:44
So let's pull up, now, something else.
47
139000
2000
Să trecem acum la altceva.
02:46
This is actually live on the Web now; you can go check it out.
48
141000
3000
Acesta este de fapt activ pe Internet acum; puteţi verifica.
02:49
This is a project called Photosynth,
49
144000
1000
Acesta este un proiect denumit Photosynth,
02:51
which really marries two different technologies.
50
146000
1000
care îmbină doua tehnologii diferite.
02:52
One of them is Seadragon
51
147000
1000
Una dintre ele este Seadragon
02:54
and the other is some very beautiful computer vision research
52
149000
2000
iar cealaltă o minunata cercetare în domeniul percepţiei vizuale computerizate
02:57
done by Noah Snavely, a graduate student at the University of Washington,
53
152000
2000
realizată de Noah Snavely, absolvent al Universităţii din Washington,
03:00
co-advised by Steve Seitz at U.W.
54
155000
2000
îndrumat de Steve Seitz la U.W.
03:02
and Rick Szeliski at Microsoft Research. A very nice collaboration.
55
157000
4000
şi Rick Szeliski la Cercetare Microsoft. O colaborare frumoasă.
03:07
And so this is live on the Web. It's powered by Seadragon.
56
162000
2000
Astfel aceasta este activă pe Internet. Posibilă datorită Seadragon.
03:09
You can see that when we kind of do these sorts of views,
57
164000
2000
Puteţi observa că atunci când creăm astfel de perspective,
03:12
where we can dive through images
58
167000
1000
când putem naviga prin imagini
03:14
and have this kind of multi-resolution experience.
59
169000
1000
putem avea acest tip de experienţă în multe rezoluţii.
03:16
But the spatial arrangement of the images here is actually meaningful.
60
171000
4000
Dar aranjamentul spaţial al acestor imagini contează cu adevărat.
03:20
The computer vision algorithms have registered these images together
61
175000
3000
Algoritmii de percepţie vizuală au înregistrat aceste imagini împreună,
03:23
so that they correspond to the real space in which these shots --
62
178000
4000
astfel încât să corespundă spaţiului real unde aceste poze --
03:27
all taken near Grassi Lakes in the Canadian Rockies --
63
182000
2000
toate facute lângă Lacurile Grassi din Stâncile Canadiene --
03:31
all these shots were taken. So you see elements here
64
186000
2000
au fost realizate. Vedeţi aceste elemente aici
03:33
of stabilized slide-show or panoramic imaging,
65
188000
4000
ale unei proiecţii stabilizate sau panoramare,
03:40
and these things have all been related spatially.
66
195000
2000
iar aceste elemente au fost relaţionate spaţial.
03:42
I'm not sure if I have time to show you any other environments.
67
197000
3000
Nu sunt sigur dacă am timp să vă arăt alte medii.
03:45
There are some that are much more spatial.
68
200000
1000
Există unele care au mult mai multa spaţialitate.
03:47
I would like to jump straight to one of Noah's original data-sets --
69
202000
3000
Aş vrea să mergem direct la unul din seturile originale ale lui Noah --
03:50
and this is from an early prototype of Photosynth
70
205000
2000
acesta face parte dintr-un prototip timpuriu al Photosynth
03:52
that we first got working in the summer --
71
207000
2000
la care am lucrat în vară --
03:54
to show you what I think
72
209000
1000
ca să vă arăt ceea ce cred eu
03:55
is really the punch line behind this technology,
73
210000
3000
că este piesa de rezistenţă a acestei tehnologii,
03:59
the Photosynth technology. And it's not necessarily so apparent
74
214000
2000
tehnologia Photosynth. Nu este neapărat evident
04:01
from looking at the environments that we've put up on the website.
75
216000
3000
doar vizionând proiecţiile pe care le-am pus la dispoziţie pe site.
04:04
We had to worry about the lawyers and so on.
76
219000
2000
A trebuit să avem grijă de aspectele legale ş.a.m.d.
04:07
This is a reconstruction of Notre Dame Cathedral
77
222000
1000
Aceaste este o reconstrucţie a Catedralei Notre Dame
04:09
that was done entirely computationally
78
224000
2000
făcută complet computaţional
04:11
from images scraped from Flickr. You just type Notre Dame into Flickr,
79
226000
3000
din imagini culese de pe Flickr. Doar cauţi Notre Dame pe Flickr,
04:14
and you get some pictures of guys in t-shirts, and of the campus
80
229000
3000
şi găseşti ceva poze cu tipi în tricou, din campus
04:17
and so on. And each of these orange cones represents an image
81
232000
4000
ş.a.m.d. Iar fiecare din aceste conuri portocalii reprezintă o imagine
04:22
that was discovered to belong to this model.
82
237000
2000
care a fost găsită că ar corespunde acestui model.
04:26
And so these are all Flickr images,
83
241000
2000
Acestea sunt poze de pe Flickr,
04:28
and they've all been related spatially in this way.
84
243000
3000
şi toate au fost relaţionate spaţial în acest mod.
04:31
And we can just navigate in this very simple way.
85
246000
2000
Putem pur şi simplu să navigăm în acest mod.
04:35
(Applause)
86
250000
9000
(Aplauze)
04:44
You know, I never thought that I'd end up working at Microsoft.
87
259000
2000
Ştiţi, nu m-am gândit niciodată că voi ajunge să lucrez la Microsoft
04:46
It's very gratifying to have this kind of reception here.
88
261000
4000
Este foarte măgulitor să am parte de o asemenea primire aici.
04:50
(Laughter)
89
265000
3000
(Râsete)
04:53
I guess you can see
90
268000
3000
Bănuiesc că observaţi
04:56
this is lots of different types of cameras:
91
271000
2000
că sunt o multitudine de camere:
04:58
it's everything from cell phone cameras to professional SLRs,
92
273000
3000
toată gama de la telefoane celulare la SLR-uri profesionale,
05:02
quite a large number of them, stitched
93
277000
1000
chiar un număr foarte mare, împletite
05:03
together in this environment.
94
278000
1000
în acest mediu.
05:04
And if I can, I'll find some of the sort of weird ones.
95
279000
2000
Dacă pot, voi mai găsi dintr-acestea ciudate.
05:08
So many of them are occluded by faces, and so on.
96
283000
3000
Sunt atât de multe mascate de feţe, ş.a.m.d.
05:13
Somewhere in here there are actually
97
288000
1000
Undeva pe aici este o serie
05:15
a series of photographs -- here we go.
98
290000
1000
de fotografii -- iată.
05:17
This is actually a poster of Notre Dame that registered correctly.
99
292000
3000
Acesta este un poster cu Notre Dame care a fost înregistrat corect.
05:21
We can dive in from the poster
100
296000
2000
Te poţi îndepărta de poster
05:24
to a physical view of this environment.
101
299000
3000
spre un mediu fizic al acestui mediu.
05:31
What the point here really is is that we can do things
102
306000
3000
Ideea este că se pot face anumite lucruri
05:34
with the social environment. This is now taking data from everybody --
103
309000
5000
cu mediul social. Acest lucru înseamnă să colectezi date de la toată lumea --
05:39
from the entire collective memory
104
314000
1000
de la o întregă memorie colectivă
05:40
of, visually, of what the Earth looks like --
105
315000
2000
despre ce însemnă Pământul vizual --
05:43
and link all of that together.
106
318000
1000
şi să le interconectezi.
05:44
All of those photos become linked together,
107
319000
2000
Toate acele poze au fost interconectate,
05:46
and they make something emergent
108
321000
1000
au devenit ceva emergent
05:47
that's greater than the sum of the parts.
109
322000
2000
care este mai mare decât suma părţilor.
05:49
You have a model that emerges of the entire Earth.
110
324000
2000
Aveţi un model al ceea ce iese din întregul Pământ.
05:51
Think of this as the long tail to Stephen Lawler's Virtual Earth work.
111
326000
5000
Gândiţi-vă la asta ca la coada lungă a lucrării lui Stephen Lawler "Pământul Virtual".
05:56
And this is something that grows in complexity
112
331000
2000
Iar aceasta creşte în complexitate
05:58
as people use it, and whose benefits become greater
113
333000
3000
pe măsură ce lumea îl foloseşte, şi ale cărui beneficii devin mai mari
06:01
to the users as they use it.
114
336000
2000
pe măsură ce utilizatorii îl folosesc.
06:03
Their own photos are getting tagged with meta-data
115
338000
2000
Propriile poze sunt etichetate cu meta-data
06:05
that somebody else entered.
116
340000
1000
pe care altcineva a introdus-o.
06:07
If somebody bothered to tag all of these saints
117
342000
3000
Dacă cineva ar eticheta toţi aceşti sfinţi
06:10
and say who they all are, then my photo of Notre Dame Cathedral
118
345000
3000
şi ar spune cine sunt, atunci poza mea cu Catedrala Notre Dame
06:13
suddenly gets enriched with all of that data,
119
348000
2000
s-ar îmbunătăţi brusc cu toate aceste date,
06:15
and I can use it as an entry point to dive into that space,
120
350000
3000
şi aş putea-o folosi ca punct de placare pentru a naviga prin acel spaţiu,
06:18
into that meta-verse, using everybody else's photos,
121
353000
2000
într-un meta-univers
06:21
and do a kind of a cross-modal
122
356000
2000
pentru a realiza astfel un fel de
06:25
and cross-user social experience that way.
123
360000
3000
experienţă socială combinată.
06:28
And of course, a by-product of all of that
124
363000
1000
Şi, bineînţeles, un produs secundar a toate acestea
06:30
is immensely rich virtual models
125
365000
2000
sunt imensele şi bogatele modele virtuale
06:32
of every interesting part of the Earth, collected
126
367000
2000
a fiecărei părţi interesante a Pământului, colecţionate
06:35
not just from overhead flights and from satellite images
127
370000
3000
nu doar din zboruri deasupra pământului şi din imagini din satelit
06:38
and so on, but from the collective memory.
128
373000
2000
ş.a.m.d., ci dintr-o memorie colectivă.
06:40
Thank you so much.
129
375000
2000
Vă mulţumesc foarte mult.
06:42
(Applause)
130
377000
11000
(Aplauze)
06:53
Chris Anderson: Do I understand this right? That what your software is going to allow,
131
388000
4000
Chris Anderson: Am înţeles corect? Ceea ce va permite softul vostru,
06:58
is that at some point, really within the next few years,
132
393000
2000
este ca la un moment dat, în urmatorii câţiva ani,
07:01
all the pictures that are shared by anyone across the world
133
396000
4000
este ca toate pozele împărtăşite de cineva de pe glob
07:05
are going to basically link together?
134
400000
2000
să fie interrelaţionate?
07:07
BAA: Yes. What this is really doing is discovering.
135
402000
2000
BAA: Da. Ceea ce face cu adevarat este să descopere.
07:09
It's creating hyperlinks, if you will, between images.
136
404000
3000
Creează hiper-legături, dacă vreţi, între imagini.
07:12
And it's doing that
137
407000
1000
Şi realizează asta
07:13
based on the content inside the images.
138
408000
1000
bazându-se pe conţinutul imaginilor.
07:14
And that gets really exciting when you think about the richness
139
409000
3000
Şi aceasta devine incitant când te gândeşti la bogăţia
07:17
of the semantic information that a lot of those images have.
140
412000
2000
informaţiei semantice pe care o au acele imagini.
07:19
Like when you do a web search for images,
141
414000
2000
Ca atunci când faci o cautare de imagini pe Internet,
07:22
you type in phrases, and the text on the web page
142
417000
2000
cauţi dupa anumite cuvinte, şi textul de pe pagina de Internet
07:24
is carrying a lot of information about what that picture is of.
143
419000
3000
dă o multitudine de informaţii despre ce reprezintă poza.
07:27
Now, what if that picture links to all of your pictures?
144
422000
2000
Acum, ce s-ar fi dacă acea poză face legătura cu toate pozele tale?
07:29
Then the amount of semantic interconnection
145
424000
2000
Atunci cantitatea de interconectare semantică
07:31
and the amount of richness that comes out of that
146
426000
1000
şi bogăţia care derivă din aceasta
07:32
is really huge. It's a classic network effect.
147
427000
3000
este imensă. Este unde efect clasic de reţea.
07:35
CA: Blaise, that is truly incredible. Congratulations.
148
430000
2000
CA: Blaise, este într-adevăr incredibil. Felicitări.
07:37
BAA: Thanks so much.
149
432000
1000
BAA: Mulţumesc foarte mult.
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