TED2007
Blaise Agüera y Arcas: How PhotoSynth can connect the world's images
Blaise Aguera y Arcas anaonyesha Photosynth
Filmed:
Readability: 4.3
5,831,957 views
Blaise Aguera y Arcas anaonyesha Photosynth, mfumo wa kompyuta wa kusanii picha ambao utabadili namna tuonavyo picha za dijito. Kwa kutumia picha zilizokusanywa kwenye Mtandao wa Intaneti, Photosynth inajenga taswira murua na kutuwezesha kuzitembelea.
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
Nitakachowaonyesha kwanza, haraka niwezavyo,
00:27
is some foundational work, some new technology
1
2000
4000
ni kazi ya msingi, katika teknolojia mpya
00:31
that we brought to Microsoft as part of an acquisition
2
6000
3000
ambayo tuliingiza Microsoft kama sehemu ya ununuzi wa kampuni
00:34
almost exactly a year ago. This is Seadragon,
3
9000
3000
takriban mwaka mmoja uliopita. Hii ni Seadragon.
00:37
and it's an environment in which you can either locally or remotely
4
12000
3000
Na ni mfumo ambao unaweza, kwa karibu au mbali,
00:40
interact with vast amounts of visual data.
5
15000
3000
kujiunganisha na kufanyia kazi takwimu mbalimbali za picha.
00:43
We're looking at many, many gigabytes of digital photos here
6
18000
3000
Hapa tunaziangalia nyingi, katika kipimo cha picha cha gigabyte
00:46
and kind of seamlessly and continuously zooming in,
7
21000
3000
na bila kukatika na kwa kuendelea kukuza mfululizo,
00:50
panning through the thing, rearranging it in any way we want.
8
25000
2000
kulengesha kwenye kitu, kuirekebisha vile tutakavyo.
00:52
And it doesn't matter how much information we're looking at,
9
27000
4000
Bila kujali taarifa ngapi tunaziangalia,
00:56
how big these collections are or how big the images are.
10
31000
3000
zina ukubwa gani au nyingi kiasi gani.
00:59
Most of them are ordinary digital camera photos,
11
34000
2000
Nyingi kati yake ni picha za kawaida zilizopigwa na kamera za dijito,
01:01
but this one, for example, is a scan from the Library of Congress,
12
36000
3000
hii hapa, kwa mfano, ni kivuli cha picha kutoka Maktaba ya Bunge,
01:05
and it's in the 300 megapixel range.
13
40000
2000
na iko katika kipimo cha vipandepicha 300.
01:08
It doesn't make any difference
14
43000
1000
Haileti tofauti yeyote
01:09
because the only thing that ought to limit the performance
15
44000
3000
kwasababu kitu pekee kitakachoweza kuzuia ufanisi
01:12
of a system like this one is the number of pixels on your screen
16
47000
3000
wa mfumo kama huu ni idadi ya vipandepicha kwenye skrini yako
01:15
at any given moment. It's also very flexible architecture.
17
50000
3000
wakati wowote. Huu ni usanifu huria.
01:18
This is an entire book, so this is an example of non-image data.
18
53000
3000
Hiki ni kitabu kizima, mfano wa takwimu ambazo si picha.
01:22
This is "Bleak House" by Dickens. Every column is a chapter.
19
57000
5000
Hiki ni Bleak House kilichoandikwa na Dickens. Kila safu ni sura.
01:27
To prove to you that it's really text, and not an image,
20
62000
4000
Kuwathibitishia kwamba haya ni maandishi, na siyo picha,
01:31
we can do something like so, to really show
21
66000
2000
tunaweza kufanya kama hivi, ili kuweza kuonyesha
01:33
that this is a real representation of the text; it's not a picture.
22
68000
3000
kuwa hiki ni kielelezo cha maandishi: na siyo picha.
01:37
Maybe this is a kind of an artificial way to read an e-book.
23
72000
2000
Labda hii ni njia nyingine ya kusoma kitabu cha nakala za elektroniki.
01:39
I wouldn't recommend it.
24
74000
1000
Siwezi kuipendekeza.
01:40
This is a more realistic case. This is an issue of The Guardian.
25
75000
3000
Huu ni mfano wa ukweli. Hili ni toleo la The Guardian.
01:43
Every large image is the beginning of a section.
26
78000
2000
Kila picha kubwa ni mwanzo wa sura.
01:45
And this really gives you the joy and the good experience
27
80000
3000
Hii inakupa raha na uzoefu mzuri
01:48
of reading the real paper version of a magazine or a newspaper,
28
83000
5000
wa kusoma tolea halisi la jarida au gazeti,
01:54
which is an inherently multi-scale kind of medium.
29
89000
1000
ambalo mara nyingi chombo cha habari chenye kina na mapana.
01:56
We've also done a little something
30
91000
1000
Pia tumefanya kitu kidogo
01:57
with the corner of this particular issue of The Guardian.
31
92000
3000
kwenye kona ya toleo hili la The Guardian.
02:00
We've made up a fake ad that's very high resolution --
32
95000
3000
Tumetengeneza tangazo la uongo na ambalo liko katika kiwango cha juu sana --
02:03
much higher than you'd be able to get in an ordinary ad --
33
98000
2000
kuliko ambavyo ungeweza kuona kwenye tangazo la kawaida --
02:05
and we've embedded extra content.
34
100000
2000
na tumeongezea vitu vya ziada.
02:07
If you want to see the features of this car, you can see it here.
35
102000
2000
Kama unataka kujua taarifa za undani wa gari hili, unaweza kuziona hapa.
02:10
Or other models, or even technical specifications.
36
105000
4000
Au miundo mingine, au hata maelezo ya kina ya kiufundi.
02:15
And this really gets at some of these ideas
37
110000
2000
Hii inaingia katika baadhi ya haya mawazo
02:18
about really doing away with those limits on screen real estate.
38
113000
4000
katika kuondokana na vikwazo vya ufanisi wa skrini
02:22
We hope that this means no more pop-ups
39
117000
2000
Tunatumaini kwamba hii ina maana kwamba hakutakuwa na vipeperushitovuti tena
02:24
and other kind of rubbish like that -- shouldn't be necessary.
40
119000
2000
na taka nyingine kama hizo -- hazitakuwa muhimu.
02:27
Of course, mapping is one of those really obvious applications
41
122000
2000
hakika, ramani zitakuwa moja ya matumizi muhimu ya
02:29
for a technology like this.
42
124000
2000
teknolojia kama hii
02:31
And this one I really won't spend any time on,
43
126000
2000
Na sitapoteza muda kwenye hili,
02:33
except to say that we have things to contribute to this field as well.
44
128000
2000
isipokuwa kwamba tuna vitu vya kuchangia katika eneo hili pia.
02:37
But those are all the roads in the U.S.
45
132000
2000
Lakini hizi zote ni barabara za Marekani
02:39
superimposed on top of a NASA geospatial image.
46
134000
4000
zilizowekwa juu ya picha za kijiografia za NASA
02:44
So let's pull up, now, something else.
47
139000
2000
Sasa hebu tuangalie kitu kingine.
02:46
This is actually live on the Web now; you can go check it out.
48
141000
3000
Hii ipo hewani kwenye mtando kwa sasa; unaweza kwenda na kuiangalia.
02:49
This is a project called Photosynth,
49
144000
1000
Huu ni mradi unaoitwa Photosynth,
02:51
which really marries two different technologies.
50
146000
1000
ambao unajumuisha teknolojia mbili tofauti.
02:52
One of them is Seadragon
51
147000
1000
Mojawapo ni Seadragon
02:54
and the other is some very beautiful computer vision research
52
149000
2000
na nyingine ni ya utafiti wa kuona katika kompyuta
02:57
done by Noah Snavely, a graduate student at the University of Washington,
53
152000
2000
uliofanywa na Noah Snavely, mwanafunzi wa chuo kikuu cha Washington,
03:00
co-advised by Steve Seitz at U.W.
54
155000
2000
na kushauriwa na Steve Seitz hapo UW
03:02
and Rick Szeliski at Microsoft Research. A very nice collaboration.
55
157000
4000
na Rick Szeliski katika kitengo cha utafiti cha Microsoft. Ushirikiano mzuri sana.
03:07
And so this is live on the Web. It's powered by Seadragon.
56
162000
2000
Kwa hiyo hii iko hewani kwenye mtandao. Na imewezeshwa na Seadragon.
03:09
You can see that when we kind of do these sorts of views,
57
164000
2000
Unaweza kuona wakati tukifanya vielelezo hivi,
03:12
where we can dive through images
58
167000
1000
ambapo tunaweza kuzamia kwenye picha
03:14
and have this kind of multi-resolution experience.
59
169000
1000
na kuwa na aina hii ya kuweza kuona taswira mbalimbali.
03:16
But the spatial arrangement of the images here is actually meaningful.
60
171000
4000
Lakini hapa mpangalio wa mahusiano ya picha unaleta maana zaidi.
03:20
The computer vision algorithms have registered these images together
61
175000
3000
Miundonamba ya picha za kompyuta imezisajiri hizi picha pamoja,
03:23
so that they correspond to the real space in which these shots --
62
178000
4000
ili ziendane na sehemu halisi ambako picha hizi zilipigwa --
03:27
all taken near Grassi Lakes in the Canadian Rockies --
63
182000
2000
zote zilipigwa karibu na Ziwa Grassi huko Canadian Rockies --
03:31
all these shots were taken. So you see elements here
64
186000
2000
zilichukuliwa. Kwa hiyo unaona vipengee hapa
03:33
of stabilized slide-show or panoramic imaging,
65
188000
4000
za vielelezopicha vilivyokamilika au picha za kupita.
03:40
and these things have all been related spatially.
66
195000
2000
na vitu hivi vyote vimehusianishwa pamoja.
03:42
I'm not sure if I have time to show you any other environments.
67
197000
3000
Sina uhakika kama nina muda wa kuwaonyesha taswira nyingine.
03:45
There are some that are much more spatial.
68
200000
1000
Kunamengine ambayo yanahusiana zaidi.
03:47
I would like to jump straight to one of Noah's original data-sets --
69
202000
3000
Nitaenda moja kwa moja kwenye moja ya seti za takwimu halisi za Noah --
03:50
and this is from an early prototype of Photosynth
70
205000
2000
na hii inatoka kwenye toleo la mfano la Photosynth ya awali
03:52
that we first got working in the summer --
71
207000
2000
ambayo tuliipata wakati tukifanya kazi majira ya joto --
03:54
to show you what I think
72
209000
1000
kukuonyesha ninachokifikiria
03:55
is really the punch line behind this technology,
73
210000
3000
ni mzaha tu wa teknolojia hii,
03:59
the Photosynth technology. And it's not necessarily so apparent
74
214000
2000
teknolojia ya Photosynth. Na si dhahiri sana
04:01
from looking at the environments that we've put up on the website.
75
216000
3000
kwa kuangalia katika mfumo tuliouweka kwenye tovuti.
04:04
We had to worry about the lawyers and so on.
76
219000
2000
Ilibidi tuanze kuhofia juu ya wanasheria na mengineyo.
04:07
This is a reconstruction of Notre Dame Cathedral
77
222000
1000
Huu ni ujengwaji tena wa kanisa kuu la dayosisi ya Notre Dame
04:09
that was done entirely computationally
78
224000
2000
ambao ulifanywa kwa kwakutumia kompyuta peke yake
04:11
from images scraped from Flickr. You just type Notre Dame into Flickr,
79
226000
3000
kutoka kwenye picha zilizopatikana kwenye Flickr. Unaandika Notre Dame kwenye Flickr,
04:14
and you get some pictures of guys in t-shirts, and of the campus
80
229000
3000
na unapata picha za watu waliovaa T-shirts, na za eneo la chuo
04:17
and so on. And each of these orange cones represents an image
81
232000
4000
na mengineyo. Na kati ya kila hizi pia za rangi ya chungwa zinawakilisha taswira
04:22
that was discovered to belong to this model.
82
237000
2000
ambazo ziligunduliwa zinauhusiano na muundo huu.
04:26
And so these are all Flickr images,
83
241000
2000
Na hizi zote ni picha za Flickr,
04:28
and they've all been related spatially in this way.
84
243000
3000
na zote zimehusishwa kwa njia hii.
04:31
And we can just navigate in this very simple way.
85
246000
2000
Na tunaweza kutembelea kwa njia hii rahisi.
04:35
(Applause)
86
250000
9000
(Makofi).
04:44
You know, I never thought that I'd end up working at Microsoft.
87
259000
2000
Unajua, sikufikiria kuwa nitakuja kufanya kazi Microsoft.
04:46
It's very gratifying to have this kind of reception here.
88
261000
4000
Ni faraja kubwa sana kupata mapokezi kama haya hapa.
04:50
(Laughter)
89
265000
3000
(Kicheko).
04:53
I guess you can see
90
268000
3000
Natumaini mnaweza kuona
04:56
this is lots of different types of cameras:
91
271000
2000
hizi ni kamera nyingi tofauti:
04:58
it's everything from cell phone cameras to professional SLRs,
92
273000
3000
ni kila kitu kutoka kwenye kamera za simu za mkononi mpaka kamera za kitaalam za SLRs,
05:02
quite a large number of them, stitched
93
277000
1000
ni nyingi, zikiwa pamoja
05:03
together in this environment.
94
278000
1000
katika mfumo huu.
05:04
And if I can, I'll find some of the sort of weird ones.
95
279000
2000
Na kama nitaweza, nitatafuta zile za ajabu.
05:08
So many of them are occluded by faces, and so on.
96
283000
3000
Nyingi zao zimezibwa kwa sura za watu, na mengineyo
05:13
Somewhere in here there are actually
97
288000
1000
Kati ya hapo kuna
05:15
a series of photographs -- here we go.
98
290000
1000
mlolongo wa picha -- naam hapa.
05:17
This is actually a poster of Notre Dame that registered correctly.
99
292000
3000
Hii hakika ni picha ya Notre Dame ambayo imesajiliwa kwa usahihi.
05:21
We can dive in from the poster
100
296000
2000
Tunaweza kuingia ndani ya picha
05:24
to a physical view of this environment.
101
299000
3000
katika mazingira ya maumbile yake.
05:31
What the point here really is is that we can do things
102
306000
3000
Cha muhimu hapa ni nini tunaweza kufanya
05:34
with the social environment. This is now taking data from everybody --
103
309000
5000
na mfumo huu. Hii ni kuchukua takwimu kutoka kwa kila mtu --
05:39
from the entire collective memory
104
314000
1000
kutoka katika mkusanyiko wa kumbukumbu
05:40
of, visually, of what the Earth looks like --
105
315000
2000
za taswira, namna dunia ilivyo --
05:43
and link all of that together.
106
318000
1000
na kuzijumuisha zote.
05:44
All of those photos become linked together,
107
319000
2000
Picha zote zinaunganishwa pamoja,
05:46
and they make something emergent
108
321000
1000
na zinafanya kitu kutokea
05:47
that's greater than the sum of the parts.
109
322000
2000
ambacho ni kubwa zaidi ya jumla ya sehemu ndogondogo.
05:49
You have a model that emerges of the entire Earth.
110
324000
2000
Una mfano ambao unatokea katika dunia nzima.
05:51
Think of this as the long tail to Stephen Lawler's Virtual Earth work.
111
326000
5000
Fikiria hii ni kama mkia mrefu wa kazi za picha za dunia za Stephen Lawler.
05:56
And this is something that grows in complexity
112
331000
2000
Na hiki kitu ambacho kinakua na kuongeza mchangamano
05:58
as people use it, and whose benefits become greater
113
333000
3000
jinsi watu wanavyotumia, na faida yake inakuwa kubwa
06:01
to the users as they use it.
114
336000
2000
kwa watumiaji jinsi wanavyotumia.
06:03
Their own photos are getting tagged with meta-data
115
338000
2000
Picha zao zinaunganishwa na meta-data
06:05
that somebody else entered.
116
340000
1000
ambavyo mtu mwingine ameviingiza.
06:07
If somebody bothered to tag all of these saints
117
342000
3000
Kama kuna mtu angewaunganisha watakatifu hawa wote
06:10
and say who they all are, then my photo of Notre Dame Cathedral
118
345000
3000
na kusema wao ni akina nani, kwa hiyo picha yangu ya kanisa kuu la Notre Dame
06:13
suddenly gets enriched with all of that data,
119
348000
2000
ingeboreshwa na vielelezo hivyo vyote,
06:15
and I can use it as an entry point to dive into that space,
120
350000
3000
na ninaweza kuitumia kama njia ya kuingia katika sehemu hiyo,
06:18
into that meta-verse, using everybody else's photos,
121
353000
2000
katika takwimumaneno, kwa kutumia picha za watu wengine,
06:21
and do a kind of a cross-modal
122
356000
2000
na kufanya mwingiliano
06:25
and cross-user social experience that way.
123
360000
3000
na mwingiliano wa watumiaji kwa njia hiyo.
06:28
And of course, a by-product of all of that
124
363000
1000
Na kwa hakika, matokeo ya yote hayo
06:30
is immensely rich virtual models
125
365000
2000
ni mifumo thabiti ya picha
06:32
of every interesting part of the Earth, collected
126
367000
2000
wa kila sehemu ya dunia, iliyokusanywa
06:35
not just from overhead flights and from satellite images
127
370000
3000
siyo tu kwa angani na kutoka kwenye picha za setilaiti
06:38
and so on, but from the collective memory.
128
373000
2000
na mengineyo, bali kutoka kwenye majumuisho ya kumbukumbu.
06:40
Thank you so much.
129
375000
2000
Asanteni sana.
06:42
(Applause)
130
377000
11000
(Makofi).
06:53
Chris Anderson: Do I understand this right? That what your software is going to allow,
131
388000
4000
Chris Anderson: Nimekuelewa? Kuwa programu yako itaruhusu,
06:58
is that at some point, really within the next few years,
132
393000
2000
kuwa, wakati fulani, katika kipindi cha miaka michache ijayo,
07:01
all the pictures that are shared by anyone across the world
133
396000
4000
picha zote zitakazokuwa zikigawanwa na mtu yeyote duniani
07:05
are going to basically link together?
134
400000
2000
zitaunganishwa pamoja?
07:07
BAA: Yes. What this is really doing is discovering.
135
402000
2000
BAA: Ndiyo. Kinachotokea hapa ni uvumbuzi.
07:09
It's creating hyperlinks, if you will, between images.
136
404000
3000
Inatengeneza viuongotovuti, kati ya picha.
07:12
And it's doing that
137
407000
1000
Na inafanya hivyo
07:13
based on the content inside the images.
138
408000
1000
ikitegemea yaliyomo ndani ya picha.
07:14
And that gets really exciting when you think about the richness
139
409000
3000
Hii inaleta msisimko zaidi ukifikiria kuhusu ubora
07:17
of the semantic information that a lot of those images have.
140
412000
2000
wa taarifa zilizomo kwenye picha hizo.
07:19
Like when you do a web search for images,
141
414000
2000
Kwa mfano ukiwa unatafuta picha kwenye mtandao,
07:22
you type in phrases, and the text on the web page
142
417000
2000
unaandika vifungu vya maneno na maandishi katika ukurasa wa tovuti
07:24
is carrying a lot of information about what that picture is of.
143
419000
3000
inabeba taarifa kuhusu picha hiyo ni ya nini.
07:27
Now, what if that picture links to all of your pictures?
144
422000
2000
Sasa, itakuwaje kama picha hiyo inaunganisha picha zako zote?
07:29
Then the amount of semantic interconnection
145
424000
2000
Hapo idadi ya miunganiko ya taarifa
07:31
and the amount of richness that comes out of that
146
426000
1000
na idadi ya ubora ambao unakuja pamoja nayo
07:32
is really huge. It's a classic network effect.
147
427000
3000
ni kubwa sana. Ni matokeo ya kiwango cha juu cha muungano wa mtandao.
07:35
CA: Blaise, that is truly incredible. Congratulations.
148
430000
2000
CA: Blaise, hii ni nzuri sana. Hongera.
07:37
BAA: Thanks so much.
149
432000
1000
BAA: Asante sana
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