Blaise Aguera y Arcas: Jaw-dropping Photosynth demo

45,610 views ใƒป 2007-06-26

TED


์•„๋ž˜ ์˜๋ฌธ์ž๋ง‰์„ ๋”๋ธ”ํด๋ฆญํ•˜์‹œ๋ฉด ์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค.

๊ฒ€ํ† : John Lynch
00:25
What I'm going to show you first, as quickly as I can,
0
25000
2548
๋จผ์ €, ๊ฐ€๋Šฅํ•œ ํ•œ ๋นจ๋ฆฌ, ๋ณด์—ฌ๋“œ๋ฆด ๊ฒƒ์€,
00:27
is some foundational work, some new technology
1
27572
3769
๊ฑฐ์˜ ์ •ํ™•ํžˆ ์ผ ๋…„ ์ „์— ํ•ฉ๋ณ‘์˜ ์ผํ™˜์œผ๋กœ
00:31
that we brought to Microsoft as part of an acquisition
2
31365
2611
์šฐ๋ฆฌ๊ฐ€ ๋งˆ์ดํฌ๋กœ์†Œํ”„ํŠธ์— ๊ฐ€์ง€๊ณ  ๊ฐ”๋˜
00:34
almost exactly a year ago.
3
34000
1821
์–ด๋–ค ๊ธฐ์ดˆ ์ž‘์—…, ์‹ ๊ธฐ์ˆ ์ž…๋‹ˆ๋‹ค.
00:35
This is Seadragon, and it's an environment
4
35845
2368
์—ฌ๋Ÿฌ๋ถ„์ด ํ˜„์ง€์—์„œ๋‚˜ ์›๊ฑฐ๋ฆฌ์—์„œ๋‚˜
00:38
in which you can either locally or remotely interact
5
38237
2476
00:40
with vast amounts of visual data.
6
40737
2119
๋ง‰๋Œ€ํ•œ ์–‘์˜ ์‹œ๊ฐ ๋ฐ์ดํ„ฐ๋กœ ์ƒํ˜ธ ์ž‘์šฉํ•  ์ˆ˜ ์žˆ๋Š” ํ™˜๊ฒฝ์ž…๋‹ˆ๋‹ค.
00:43
We're looking at many, many gigabytes of digital photos here
7
43165
3404
์—ฌ๊ธฐ์„œ ์šฐ๋ฆฌ๋Š” ์ƒ๋‹นํžˆ ๋งŽ์€ ๊ธฐ๊ฐ€๋ฐ”์ดํŠธ์˜ ๋””์ง€ํ„ธ ์‚ฌ์ง„๋“ค์„ ๋ณด๊ณ  ์žˆ๊ณ ,
00:46
and kind of seamlessly and continuously zooming in,
8
46593
2915
์ด์Œ์ƒˆ์—†์ด ์•„์ฃผ ๋งค๋„๋Ÿฝ๊ณ  ์—ฐ์†์ ์œผ๋กœ ์คŒ ์ธํ•ด ๋“ค์–ด๊ฐ€๊ณ ,
00:49
panning through it, rearranging it in any way we want.
9
49532
2545
๋‹ค๋ฅธ ์ชฝ์œผ๋กœ ํŒจ๋‹ํ•˜๊ณ , ์šฐ๋ฆฌ๊ฐ€ ์›ํ•˜๋Š” ์–ด๋–ค ์‹์œผ๋กœ๋“  ์žฌ๋ฐฐ์น˜๋ฅผ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
00:52
And it doesn't matter how much information we're looking at,
10
52389
3587
์šฐ๋ฆฌ๊ฐ€ ์–ผ๋งˆ๋‚˜ ๋งŽ์€ ์ •๋ณด๋ฅผ ๋ณด๊ณ  ์žˆ๋Š”์ง€,
00:56
how big these collections are or how big the images are.
11
56000
2976
์ด ์ปฌ๋ ‰์…˜์ด ์ •๋ง ์–ผ๋งˆ๋‚˜ ํฐ ์ง€, ์ด๋ฏธ์ง€๋“ค์ด ์–ผ๋งˆ๋‚˜ ํฐ ์ง€๋Š” ๊ทธ๋ฆฌ ์ค‘์š”ํ•˜์ง€ ์•Š์Šต๋‹ˆ๋‹ค.
00:59
Most of them are ordinary digital camera photos,
12
59000
2286
์ด์ค‘ ๋Œ€๋ถ€๋ถ„์€ ํ‰๋ฒ”ํ•œ ๋””์ง€ํ„ธ ์นด๋ฉ”๋ผ ์‚ฌ์ง„๋“ค์ž…๋‹ˆ๋‹ค.
01:01
but this one, for example, is a scan from the Library of Congress,
13
61310
3144
ํ•˜์ง€๋งŒ, ์˜ˆ๋ฅผ ๋“ค์–ด, ์ด๊ฒƒ์€ ๊ตญํšŒ๋„์„œ๊ด€์—์„œ ์Šค์บ”ํ•œ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
01:04
and it's in the 300 megapixel range.
14
64478
2818
๊ฑฐ์˜ 300 ๋ฉ”๊ฐ€ํ”ฝ์…€ ์งœ๋ฆฌ์ž…๋‹ˆ๋‹ค.
01:07
It doesn't make any difference
15
67320
1656
๊ทธ๋ž˜๋„ ๋ณ„ ์ฐจ์ด๊ฐ€ ์—†์Šต๋‹ˆ๋‹ค.
01:09
because the only thing that ought to limit the performance of a system like this one
16
69000
4144
์ด์™€ ๊ฐ™์€ ์‹œ์Šคํ…œ์˜ ์„ฑ๋Šฅ์„ ์ œํ•œํ•˜๋Š” ์œ ์ผํ•œ ์š”์†Œ๋Š”
๊ทธ ์ˆœ๊ฐ„ ํ™”๋ฉด์˜ ํ”ฝ์…€์ˆ˜์ž…๋‹ˆ๋‹ค.
01:13
is the number of pixels on your screen at any given moment.
17
73168
2777
01:15
It's also very flexible architecture.
18
75969
1970
๋˜ํ•œ ์ด๊ฒƒ์€ ์•„์ฃผ ์œ ์—ฐํ•œ ์•„ํ‚คํ…์ฒ˜์ž…๋‹ˆ๋‹ค.
01:17
This is an entire book, so this is an example of non-image data.
19
77963
3727
์ด๊ฒƒ์€ ๋น„์ด๋ฏธ์ง€ ๋ฐ์ดํ„ฐ์˜ ์˜ˆ๋กœ์„œ, ์ฑ… ์ „์ฒด์ž…๋‹ˆ๋‹ค.
01:21
This is "Bleak House" by Dickens.
20
81714
2787
์ด๊ฒƒ์€ ๋””ํ‚จ์ฆˆ์˜ "ํ™ฉ๋Ÿ‰ํ•œ ์ง‘"์ž…๋‹ˆ๋‹ค. ๊ฐ ์—ด์ด ํ•œ ์ฑ•ํ„ฐ์ž…๋‹ˆ๋‹ค.
01:24
Every column is a chapter.
21
84525
2784
01:27
To prove to you that it's really text, and not an image,
22
87333
3643
์ด๊ฒŒ ์ด๋ฏธ์ง€๊ฐ€ ์•„๋‹Œ ์‹ค์ œ ํ…์ŠคํŠธ๋ผ๋Š” ๊ฑธ ์ฆ๋ช…ํ•ด ๋“œ๋ฆฌ๊ธฐ ์œ„ํ•ด
01:31
we can do something like so, to really show
23
91000
2048
์ด๋Ÿฐ ์ž‘์—…๋„ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด๊ฒŒ ๊ทธ๋ฆผ์ด ์•„๋‹ˆ๊ณ 
01:33
that this is a real representation of the text; it's not a picture.
24
93072
3192
์ •๋ง๋กœ ํ…์ŠคํŠธ์˜ ํ‘œํ˜„์ด๋ผ๋Š” ๊ฒƒ์„ ๋ณด์—ฌ๋“œ๋ฆฌ๊ธฐ ์œ„ํ•ด์„œ ๋ง์ด์ฃ .
01:36
Maybe this is an artificial way to read an e-book.
25
96288
2664
์–ด์ฉŒ๋ฉด ์ด๊ฒƒ์€ e-๋ถ์„ ์ฝ๋Š” ์ผ์ข…์˜ ์ธ์œ„์ ์ธ ๋ฐฉ๋ฒ•์ด๋ผ๊ณ  ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
01:38
I wouldn't recommend it.
26
98976
1200
์ถ”์ฒœํ•ด ๋“œ๋ฆฌ์ง€๋Š” ์•Š๊ฒ ์Šต๋‹ˆ๋‹ค.
01:40
This is a more realistic case, an issue of The Guardian.
27
100200
2848
์ด๊ฒƒ์€ ์ข€๋” ํ˜„์‹ค์ ์ธ ๊ฒฝ์šฐ์ž…๋‹ˆ๋‹ค. ์ด๊ฑด ๊ฐ€๋””์–ธ ์žก์ง€์ž…๋‹ˆ๋‹ค.
01:43
Every large image is the beginning of a section.
28
103072
2286
ํฐ ์ด๋ฏธ์ง€๋“ค์€ ์„น์…˜์˜ ์‹œ์ž‘์ž…๋‹ˆ๋‹ค.
01:45
And this really gives you the joy and the good experience
29
105382
2904
์ด๊ฒƒ์€ ์ •๋ง ์—ฌ๋Ÿฌ๋ถ„๊ป˜ ์ง„์งœ ์ข…์ด๋กœ ๋œ ์žก์ง€๋‚˜ ์‹ ๋ฌธ์„ ์ฝ๋Š”
01:48
of reading the real paper version of a magazine or a newspaper,
30
108310
5183
๊ธฐ์จ๊ณผ ์ข‹์€ ๊ฒฝํ—˜์„ ์ „ํ•ด๋“œ๋ฆฝ๋‹ˆ๋‹ค.
01:53
which is an inherently multi-scale kind of medium.
31
113517
2435
์›๋ž˜๊ฐ€ ๋‹ค์ค‘์Šค์ผ€์ผ ์œ ํ˜•์˜ ๋งค์ฒด๊ฑฐ๋“ ์š”.
01:55
We've done something
32
115976
1000
๋˜ํ•œ ๋ฐ”๋กœ ์ด๋ฒˆํ˜ธ์˜ ๊ฐ€๋””์–ธ ๊ตฌ์„์—
01:57
with the corner of this particular issue of The Guardian.
33
117000
2976
์ž‘์€ ์ž‘์—… ํ•˜๋‚˜๋ฅผ ํ•ด ๋†“์•˜์Šต๋‹ˆ๋‹ค.
02:00
We've made up a fake ad that's very high resolution --
34
120000
2976
์•„์ฃผ ๊ณ ํ•ด์ƒ๋„๋กœ ์œ„์กฐ ๊ด‘๊ณ ๋ฅผ ๋งŒ๋“ค์—ˆ์ง€์š”--
02:03
much higher than in an ordinary ad --
35
123000
2198
๋ณดํ†ต ๊ด‘๊ณ ์—์„œ ๋ณด์‹ค ์ˆ˜ ์žˆ๋Š” ๊ฒƒ๋ณด๋‹ค ํ›จ์”ฌ ๋” ๊ณ ํ•ด์ƒ๋„๋กœ--
02:05
and we've embedded extra content.
36
125222
1754
๊ทธ๋ฆฌ๊ณ  ์ถ”๊ฐ€ ์ปจํ…์ธ ๋ฅผ ๋‚ด์žฅํ•ด ๋„ฃ์—ˆ์Šต๋‹ˆ๋‹ค.
02:07
If you want to see the features of this car, you can see it here.
37
127000
3048
์ด ์ž๋™์ฐจ์˜ ํŠน์ง•์„ ๋ณด๊ณ  ์‹ถ์œผ์‹œ๋ฉด, ์—ฌ๊ธฐ์„œ ๋ณด์‹ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
02:10
Or other models, or even technical specifications.
38
130072
4180
์•„๋‹ˆ๋ฉด ๋‹ค๋ฅธ ๋ชจ๋ธ๋“ค, ๋˜๋Š” ๊ธฐ์ˆ  ์‚ฌ์–‘๊นŒ์ง€๋„.
02:14
And this really gets at some of these ideas
39
134276
3315
์ธํ„ฐ๋„ท ๋ถ€๋™์‚ฐ์—์„œ ์ •๋ง ๊ทธ ํ•œ๊ณ„๋ฅผ ์—†์• ๋Š” ๊ฒƒ์— ๋Œ€ํ•ด
02:17
about really doing away with those limits on screen real estate.
40
137615
4661
์ด๋Ÿฐ ์•„์ด๋””์–ด๋ฅผ ์–ป์Šต๋‹ˆ๋‹ค.
02:22
We hope that this means no more pop-ups
41
142300
2111
์ด๊ฒƒ์œผ๋กœ ์ธํ•ด ๋”์ด์ƒ ํŒ์—…์ฐฝ์ด๋‚˜
02:24
and other rubbish like that -- shouldn't be necessary.
42
144435
2541
์ด๋Ÿฐ ์ €๋Ÿฐ ์“ธ๋ฐ์—†๋Š” ๊ฒƒ๋“ค์ด ๋” ๋œจ์ง€ ์•Š๊ธฐ๋ฅผ ๋ฐ”๋ž๋‹ˆ๋‹ค. ํ•„์š”์—†์œผ๋‹ˆ๊นŒ์š”.
02:27
Of course, mapping is one of those obvious applications
43
147000
2658
๋ฌผ๋ก , ์ด๋Ÿฐ ๊ธฐ์ˆ ์—๋Š” ์ง€๋„ ์ž‘์—…์ด ๊ฐ€์žฅ ํ™•์‹คํ•œ
02:29
for a technology like this.
44
149682
1294
์‘์šฉ ํ”„๋กœ๊ทธ๋žจ ์ค‘ ํ•˜๋‚˜์ž…๋‹ˆ๋‹ค.
02:31
And this one I really won't spend any time on,
45
151000
2191
๊ทธ๋ฆฌ๊ณ  ์—ฌ๊ธฐ์—๋Š” ์ •๋ง ์‹œ๊ฐ„์„ ์“ฐ์ง€ ์•Š๊ฒ ์Šต๋‹ˆ๋‹ค.
02:33
except to say that we have things to contribute to this field as well.
46
153215
3334
๋‹จ, ์ด ๋ง์”€๋งŒ ๋“œ๋ฆฌ๊ณ  ์‹ถ์–ด์š”. ์ €ํฌ๋Š” ์ด ๋ถ„์•ผ์— ๊ธฐ์—ฌํ•  ๋งŒํ•œ ๊ฒƒ๋„ ๊ฐ€์ง€๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
02:37
But those are all the roads in the U.S.
47
157213
1858
์ €๊ฒƒ๋“ค์€ ๋ฏธ๊ตญ์˜ ๋ชจ๋“  ๊ธธ๋“ค์„
02:39
superimposed on top of a NASA geospatial image.
48
159095
4565
NASA ์ง€๋ฆฌ๊ณต๊ฐ„ ์ž๋ฃŒ ์œ„์— ๊ฒน์นœ ์ด๋ฏธ์ง€์ž…๋‹ˆ๋‹ค.
02:44
So let's pull up, now, something else.
49
164000
1976
์ด์ œ ์„ธ์›Œ๋†“๊ณ , ๋‹ค๋ฅธ ๊ฒƒ์„ ๋ณผ๊นŒ์š”?
02:46
This is actually live on the Web now; you can go check it out.
50
166000
2976
์ง€๊ธˆ ์›น์—์„œ ๋ผ์ด๋ธŒ ์ค‘์ด๋‹ˆ, ๊ฐ€์„œ ํ™•์ธํ•ด ๋ณด์‹ค ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
02:49
This is a project called Photosynth, which marries two different technologies.
51
169000
3704
์ด๊ฒƒ์€ ํฌํ† ์‹ ์Šค๋ผ๊ณ  ๋ถ€๋ฅด๋Š” ํ”„๋กœ์ ํŠธ์ž…๋‹ˆ๋‹ค.
๋‘ ๊ฐ€์ง€ ๋‹ค๋ฅธ ๊ธฐ์ˆ ๋“ค์ด ์ •๋ง ํ•ฉ์ณ์ง‘๋‹ˆ๋‹ค.
02:52
One of them is Seadragon
52
172728
1248
๊ทธ ์ค‘ ํ•˜๋‚˜๋Š” ์”จ๋“œ๋ž˜๊ณค์ž…๋‹ˆ๋‹ค.
02:54
and the other is some very beautiful computer-vision research
53
174000
2906
๋‹ค๋ฅธ ๊ฒƒ์€ ์•„์ฃผ ์•„๋ฆ„๋‹ค์šด ์ปดํ“จํ„ฐ ๋น„์ „ ์—ฐ๊ตฌ์ž…๋‹ˆ๋‹ค.
02:56
done by Noah Snavely, a graduate student at the University of Washington,
54
176930
3462
์›Œ์‹ฑํ„ด๋Œ€ํ•™๊ต ๋Œ€ํ•™์›์ƒ์ธ ๋…ธ์•„ ์Šค๋„ค์ด๋ธ”๋ฆฌ๊ฐ€ ๋งŒ๋“  ๊ฒƒ์ธ๋ฐ,
03:00
co-advised by Steve Seitz at U.W.
55
180416
1829
์›Œ์‹ฑํ„ด๋Œ€ํ•™๊ต์˜ ์Šคํ‹ฐ๋ธŒ ์Šคํƒ€์ด์ธ ์™€
03:02
and Rick Szeliski at Microsoft Research.
56
182269
1978
๋งˆ์ดํฌ๋กœ์†Œํ”„ํŠธ ๋ฆฌ์„œ์น˜์˜ ๋ฆญ ์ฒผ๋ฆฌ์Šคํ‚ค์˜ ๊ณต๋™ ์ง€๋„๋ฅผ ๋ฐ›์•˜์ง€์š”. ์ฐธ ๋ฉ‹์ง„ ํ˜‘๋ ฅ์ž…๋‹ˆ๋‹ค.
03:04
A very nice collaboration.
57
184271
1733
03:06
And so this is live on the Web. It's powered by Seadragon.
58
186412
3108
๊ทธ๋ฆฌ๊ณ  ์›น์—์„œ ๋ผ์ด๋ธŒ๋กœ ๋ณผ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์”จ๋“œ๋ž˜๊ณค์ด ์ง€์›ํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
03:09
You can see that when we do these sorts of views,
59
189544
2504
์šฐ๋ฆฌ๋Š” ์ด๋Ÿฐ ์ข…๋ฅ˜์˜ ๊ฒƒ๋“ค์„ ๋ณผ ๋•Œ
03:12
where we can dive through images
60
192072
1723
์ด๋ฏธ์ง€ ์†์œผ๋กœ ๊นŠ์ด ๋“ค์–ด๊ฐˆ ์ˆ˜ ์žˆ๊ณ 
03:13
and have this kind of multi-resolution experience.
61
193819
2334
์ด๋Ÿฐ ์ข…๋ฅ˜์˜ ๋‹ค์ค‘ ํ•ด์ƒ๋„ ๊ฒฝํ—˜์„ ๊ฐ–๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
03:16
But the spatial arrangement of the images here is actually meaningful.
62
196177
3799
๊ทธ๋Ÿฌ๋‚˜ ์—ฌ๊ธฐ์„œ ์ด๋ฏธ์ง€๋“ค์˜ ๊ณต๊ฐ„์ ์ธ ๋ฐฐ์น˜๊ฐ€ ์‚ฌ์‹ค ์˜๋ฏธ์žˆ์Šต๋‹ˆ๋‹ค.
03:20
The computer vision algorithms have registered these images together
63
200000
3191
์ปดํ“จํ„ฐ ๋น„์ „ ์•Œ๊ณ ๋ฆฌ๋“ฌ์€ ์ด ์ด๋ฏธ์ง€๋“ค์„ ํ•จ๊ป˜ ๋“ฑ๋กํ–ˆ๊ณ ,
03:23
so that they correspond to the real space in which these shots --
64
203215
3761
๊ทธ๋ž˜์„œ ์ด ์ด๋ฏธ์ง€๋“ค์€ ์บ๋‚˜๋‹ค ๋กํ‚ค์‚ฐ๋งฅ์˜ ๊ทธ๋ž˜์‹œ ํ˜ธ์ˆ˜ ๊ทผ์ฒ˜--
03:27
all taken near Grassi Lakes in the Canadian Rockies --
65
207000
3300
์ด ์‚ฌ์ง„๋“ค์ด ์ฐํžŒ ์ง„์งœ ๊ณต๊ฐ„๊ณผ ์ผ์น˜ํ•ฉ๋‹ˆ๋‹ค.
03:30
all these shots were taken.
66
210324
1663
์—ฌ๊ธฐ์„œ ์•ˆ์ •ํ™”๋œ ์Šฌ๋ผ์ด๋“œ ์‡ผ๋‚˜
03:32
So you see elements here
67
212011
1467
03:33
of stabilized slide-show or panoramic imaging,
68
213502
6013
์ „๋ฐฉ์œ„ ์˜์ƒ ์‹œ์Šคํ…œ์˜ ์š”์†Œ๋“ค,
03:39
and these things have all been related spatially.
69
219539
2437
์ด๊ฒƒ๋“ค์€ ๋ชจ๋‘ ๊ณต๊ฐ„์ ์œผ๋กœ ์—ฐ๊ฒฐ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค.
03:42
I'm not sure if I have time to show you any other environments.
70
222000
3000
๋‹ค๋ฅธ ํ™˜๊ฒฝ๋“ค์„ ๋ณด์—ฌ๋“œ๋ฆด ์‹œ๊ฐ„์ด ๋  ์ง€ ์ž˜ ๋ชจ๋ฅด๊ฒ ์Šต๋‹ˆ๋‹ค.
03:45
Some are much more spatial.
71
225024
1431
ํ›จ์”ฌ ๋” ๊ณต๊ฐ„์ ์ธ ๊ฒƒ๋“ค์ด ์žˆ์Šต๋‹ˆ๋‹ค.
03:46
I would like to jump straight to one of Noah's original data-sets --
72
226479
3945
๋ง‰๋ฐ”๋กœ ๋…ธ์•„์˜ ์›๋ž˜ ๋ฐ์ดํ„ฐ ์„ธํŠธ ์ค‘ ํ•˜๋‚˜์— ๋Œ€ํ•ด ์–˜๊ธฐํ• ๊นŒ ํ•ฉ๋‹ˆ๋‹ค.
03:50
this is from an early prototype that we first got working this summer --
73
230448
3552
์ด๊ฒƒ์€ ํฌํ† ์‹ ์Šค์˜ ์ดˆ๊ธฐ ์›ํ˜•์—์„œ ๋น„๋กฏ๋œ ๊ฒƒ์ธ๋ฐ,
์—ฌ๋ฆ„์— ์šฐ๋ฆฌ๊ฐ€ ์ฒ˜์Œ ์ž‘์—…ํ•˜๊ฒŒ ๋˜์—ˆ๋˜ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
03:54
to show you what I think
74
234024
1894
์ด๊ฒŒ ๋ฐ”๋กœ ์ด ๊ธฐ์ˆ , ํฌํ† ์‹ ์Šค ๊ธฐ์ˆ  ๋’ค์— ์ˆจ์–ด ์žˆ๋Š”
03:55
is really the punch line behind the Photosynth technology,
75
235942
3838
ํ•ต์‹ฌ์„ ์ฐŒ๋ฅด๋Š” ๊ฒƒ์ด๋ผ๊ณ  ์ƒ๊ฐํ•ฉ๋‹ˆ๋‹ค. ์ด๊ฒƒ์„ ๋ณด์—ฌ๋“œ๋ฆฌ๊ณ  ์‹ถ์Šต๋‹ˆ๋‹ค.
03:59
It's not necessarily so apparent
76
239804
1561
์šฐ๋ฆฌ๊ฐ€ ์›น์‚ฌ์ดํŠธ์— ์˜ฌ๋ ค๋†“๋Š” ํ™˜๊ฒฝ์„ ๋ณผ ๋•Œ
04:01
from looking at the environments we've put up on the website.
77
241389
2895
๊ทธ๋ ‡๊ฒŒ ๋ถ„๋ช…ํ•  ํ•„์š”๋Š” ์—†์Šต๋‹ˆ๋‹ค.
04:04
We had to worry about the lawyers and so on.
78
244308
2177
์šฐ๋ฆฌ๋Š” ๋ณ€ํ˜ธ์‚ฌ ๋“ฑ์— ๋Œ€ํ•ด ๊ฑฑ์ •ํ•ด์•ผ ํ–ˆ์Šต๋‹ˆ๋‹ค.
04:06
This is a reconstruction of Notre Dame Cathedral
79
246509
2301
์ด๊ฒƒ์€ ๋…ธํ‹€๋‹ด ๋Œ€์„ฑ๋‹น์„
04:08
that was done entirely computationally from images scraped from Flickr.
80
248834
3457
ํ”Œ๋ฆญ์ปค์—์„œ ์Šคํฌ๋žฉํ•œ ์ด๋ฏธ์ง€๋“ค์„ ๊ฐ€์ง€๊ณ 
์ „์ ์œผ๋กœ ์ˆ˜ํ•™์ ์œผ๋กœ ์žฌ๊ตฌ์„ฑํ•œ ๊ฒƒ์ž…๋‹ˆ๋‹ค. ํ”Œ๋ฆญ์ปค์— ๋…ธํ‹€๋‹ด์ด๋ผ๊ณ ๋งŒ ์น˜๋ฉด,
04:12
You just type Notre Dame into Flickr,
81
252315
2019
04:14
and you get some pictures of guys in T-shirts, and of the campus and so on.
82
254358
3854
ํ‹ฐ์…”์ธ  ์ž…์€ ์‚ฌ๋žŒ๋“ค, ์บ ํผ์Šค์˜ ์‚ฌ๋žŒ๋“ค ์‚ฌ์ง„ ๋“ฑ์ด ๋‚˜์˜ต๋‹ˆ๋‹ค.
์ด ์˜ค๋ Œ์ง€ ์›๋ฟ” ํ•˜๋‚˜ ํ•˜๋‚˜๊ฐ€ ์ด ๋ชจ๋ธ์— ์†ํ•˜๋Š” ๊ฒƒ์œผ๋กœ ๋ฐœ๊ฒฌ๋œ
04:18
And each of these orange cones represents an image
83
258236
3146
04:21
that was discovered to belong to this model.
84
261406
3234
์ด๋ฏธ์ง€๋ฅผ ๋‚˜ํƒ€๋ƒ…๋‹ˆ๋‹ค.
04:26
And so these are all Flickr images,
85
266000
1976
๊ทธ๋ฆฌ๊ณ  ์ด๊ฒƒ๋“ค์€ ๋ชจ๋‘ ํ”Œ๋ฆญ์ปค ์ด๋ฏธ์ง€์ž…๋‹ˆ๋‹ค.
04:28
and they've all been related spatially in this way.
86
268000
2976
๋ชจ๋‘ ์ด๋Ÿฐ ์‹์œผ๋กœ ์„œ๋กœ ๊ณต๊ฐ„์ ์œผ๋กœ ๊ด€๋ จ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค.
04:31
We can just navigate in this very simple way.
87
271000
2334
์ด๋ ‡๊ฒŒ ์•„์ฃผ ๊ฐ„๋‹จํ•œ ๋ฐฉ๋ฒ•์œผ๋กœ ๋„ค๋น„๊ฒŒ์ด์…˜์„ ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
04:35
(Applause)
88
275000
3920
(๋ฐ•์ˆ˜).
04:42
(Applause ends)
89
282557
1014
04:43
You know, I never thought that I'd end up working at Microsoft.
90
283595
2954
์ €๋Š” ์ œ๊ฐ€ ๊ฒฐ๊ตญ ๋งˆ์ดํฌ๋กœ์†Œํ”„ํŠธ์—์„œ ์ผํ•˜๊ฒŒ ๋  ๊ฒƒ์ด๋ผ๊ณ ๋Š” ์ƒ๊ฐํ•ด ๋ณธ ์ ์ด ์—†์Šต๋‹ˆ๋‹ค.
04:46
It's very gratifying to have this kind of reception here.
91
286573
3000
์—ฌ๊ธฐ์„œ ์ด๋Ÿฐ ์‹์˜ ์ ‘๋Œ€๋ฅผ ๋ฐ›๋‹ค๋‹ˆ ๋„ˆ๋ฌด๋‚˜ ๊ฐ์‚ฌํ•ฉ๋‹ˆ๋‹ค.
04:49
(Laughter)
92
289597
3379
(์›ƒ์Œ).
04:53
I guess you can see this is lots of different types of cameras:
93
293000
5048
์—ฌ๋Ÿฌ๋ถ„์€ ์ˆ˜๋งŽ์€ ์ข…๋ฅ˜์˜ ๋งŽ์€ ์นด๋ฉ”๋ผ๋“ค์„
๋ณด์‹ค ์ˆ˜ ์žˆ์„ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
04:58
it's everything from cell-phone cameras to professional SLRs,
94
298072
3161
ํ•ธ๋“œํฐ ์นด๋ฉ”๋ผ์—์„œ๋ถ€ํ„ฐ ์ „๋ฌธ๊ฐ€์šฉ SLR์— ์ด๋ฅด๊ธฐ๊นŒ์ง€ ๋‹ค์–‘ํ•ฉ๋‹ˆ๋‹ค.
05:01
quite a large number of them, stitched together in this environment.
95
301257
3191
๊ฝค ๋งŽ์€ ์ˆ˜์˜ ์‚ฌ์ง„๋“ค์„ ์ด ํ™˜๊ฒฝ์—
ํ•จ๊ป˜ ์งœ๋„ฃ์—ˆ์Šต๋‹ˆ๋‹ค.
05:04
If I can find some of the sort of weird ones --
96
304472
2632
๊ฐ€๋Šฅํ•˜๋‹ค๋ฉด, ์ €๋Š” ์ด์ƒํ•œ ์ข…๋ฅ˜์˜ ๊ฒƒ๋“ค์„ ์ฐพ์„ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
05:08
So many of them are occluded by faces, and so on.
97
308000
3322
๋งŽ์€ ๊ฒƒ๋“ค์ด ์–ผ๊ตด๋กœ ๊ฐ€๋ ค์ ธ ์žˆ์Šต๋‹ˆ๋‹ค.
05:12
Somewhere in here there is actually a series of photographs -- here we go.
98
312595
4277
์—ฌ๊ธฐ ์ €๊ธฐ๋Š” ์‚ฌ์‹ค
์ผ๋ จ์˜ ์‚ฌ์ง„๋“ค์ž…๋‹ˆ๋‹ค - ์—ฌ๊ธฐ ์žˆ๋„ค์š”.
05:16
This is actually a poster of Notre Dame that registered correctly.
99
316896
3301
์ด๊ฒƒ์ด ์‹ค์ œ๋กœ ์ •ํ™•ํ•˜๊ฒŒ ๋“ฑ๋ก๋œ ๋…ธํ‹€๋‹ด ํฌ์Šคํ„ฐ์ž…๋‹ˆ๋‹ค.
05:20
We can dive in from the poster
100
320221
3216
์šฐ๋ฆฌ๋Š” ํฌ์Šคํ„ฐ๋กœ๋ถ€ํ„ฐ
05:23
to a physical view of this environment.
101
323461
3810
์ด ํ™˜๊ฒฝ์˜ ๋ฌผ๋ฆฌ์ ์ธ ๊ด‘๊ฒฝ์œผ๋กœ ๊นŠ์ด ์ž ์ˆ˜ํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
05:31
What the point here really is
102
331421
1866
์—ฌ๊ธฐ์„œ ์ง„์งœ ์ค‘์š”ํ•œ ๊ฒƒ์€ ์šฐ๋ฆฌ๊ฐ€ ์‚ฌํšŒ์ ์ธ ํ™˜๊ฒฝ์œผ๋กœ๋„ ์ด๋Ÿฐ ๊ฒƒ๋“ค์„
05:33
is that we can do things with the social environment.
103
333311
2591
ํ•  ์ˆ˜ ์žˆ๋‹ค๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค. ์ด๊ฒƒ์€ ๋ชจ๋“  ์‚ฌ๋žŒ์œผ๋กœ๋ถ€ํ„ฐ ๋ฐ์ดํ„ฐ๋ฅผ ์ทจํ•˜๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
05:35
This is now taking data from everybody --
104
335926
3002
05:38
from the entire collective memory, visually, of what the Earth looks like --
105
338952
3871
์‹œ๊ฐ์ ์œผ๋กœ, ์ง€๊ตฌ์ฒ˜๋Ÿผ ๋ณด์ด๋Š” ๊ฒƒ์˜
์ „์ฒด์ ์ธ ์ง‘๋‹จ ๊ธฐ์–ต์œผ๋กœ๋ถ€ํ„ฐ--
05:42
and link all of that together.
106
342847
1749
๊ทธ๋ฆฌ๊ณ  ๋ชจ๋“  ๊ฒƒ๋“ค์„ ์—ฐ๊ฒฐ์‹œํ‚ต๋‹ˆ๋‹ค.
05:44
Those photos become linked, and they make something emergent
107
344620
2839
์ด ๋ชจ๋“  ์‚ฌ์ง„๋“ค์ด ์„œ๋กœ ์—ฐ๊ฒฐ๋˜๊ณ ,
๋ถ€๋ถ„๋“ค์˜ ์ดํ•ฉ๋ณด๋‹ค ํ›จ์”ฌ ํฐ
05:47
that's greater than the sum of the parts.
108
347483
1953
๊ฒƒ์ด ๋‚˜ํƒ€๋‚˜๋„๋ก ํ•ฉ๋‹ˆ๋‹ค.
05:49
You have a model that emerges of the entire Earth.
109
349460
2356
์ง€๊ตฌ ์ „์ฒด์—์„œ ๋‚˜ํƒ€๋‚˜๋Š” ๋ชจ๋ธ์ž…๋‹ˆ๋‹ค.
05:51
Think of this as the long tail to Stephen Lawler's Virtual Earth work.
110
351840
4077
์ด๊ฒƒ์„ ์Šคํ‹ฐ๋ธ ๋กค๋Ÿฌ์˜ ์‹œ๊ฐ์  ์ง€๊ตฌ ์ž‘ํ’ˆ์˜ ๊ธด ๊ผฌ๋ฆฌ๋ผ๊ณ  ์ƒ๊ฐํ•˜์„ธ์š”.
05:55
And this is something that grows in complexity as people use it,
111
355941
3200
์‚ฌ๋žŒ๋“ค์ด ์ด์šฉํ•  ๋•Œ, ๊ทธ ๋ณตํ•ฉ์„ฑ์ด
๋” ์ปค์ง€๋Š” ๊ฒƒ์ด๊ณ , ์‚ฌ์šฉ์ž๋“ค์ด ์‚ฌ์šฉํ•  ๋•Œ,
05:59
and whose benefits become greater to the users as they use it.
112
359165
3811
ํ˜œํƒ์ด ๋” ์ปค์ง€๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
06:03
Their own photos are getting tagged with meta-data that somebody else entered.
113
363000
3692
์ž๊ธฐ๊ฐ€ ์ฐ์€ ์‚ฌ์ง„๋“ค์ด ๋‹ค๋ฅธ ๋ˆ„๊ตฐ๊ฐ€๊ฐ€ ์ž…๋ ฅํ•œ
๋ฉ”ํƒ€ ๋ฐ์ดํƒ€๋ผ๋Š” ํƒœ๊ทธ๋ฅผ ๊ฐ–๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
06:06
If somebody bothered to tag all of these saints
114
366716
3360
๋ˆ„๊ตฐ๊ฐ€๊ฐ€ ์ด ๋ชจ๋“  ์„ฑ์ธ๋“ค์—๊ฒŒ ํƒœ๊ทธ๋ฅผ ๋ถ™์ด๋Š” ๊ฒƒ์ด ๊ท€์ฐฎ๊ณ ,
06:10
and say who they all are, then my photo of Notre Dame Cathedral
115
370100
2953
๋„๋Œ€์ฒด ์ด๊ฒŒ ๋‹ค ๋ˆ„๊ตฌ๋ƒ๊ณ  ํ•œ๋‹ค๋ฉด, ์ œ ์‚ฌ์ง„ ๋…ธํ‹€๋‹ด ๋Œ€์„ฑ๋‹น์ด
06:13
suddenly gets enriched with all of that data,
116
373077
2098
๊ฐ‘์ž๊ธฐ ๋‚˜ํƒ€๋‚˜ ๊ทธ ๋ชจ๋“  ๋ฐ์ดํ„ฐ๋ฅผ ํ’์„ฑํ•˜๊ฒŒ ๋งŒ๋“ค๊ณ 
06:15
and I can use it as an entry point to dive into that space,
117
375199
2777
์ €๋Š” ๊ทธ ๊ณต๊ฐ„ ๊นŠ์€ ๊ณณ์œผ๋กœ,
06:18
into that meta-verse, using everybody else's photos,
118
378000
2681
๋ฉ”ํƒ€-์‹œ ๊นŠ์€ ๊ณณ์œผ๋กœ, ๋‹ค๋ฅธ ์‚ฌ๋žŒ์˜ ์‚ฌ์ง„๋“ค์„ ์ด์šฉํ•˜์—ฌ,
06:20
and do a kind of a cross-modal
119
380705
3301
๊นŠ์ด ์ž ์ˆ˜ํ•ด ๊ฐˆ ์‹œ์ž‘์ ์œผ๋กœ ์ด์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.
06:24
and cross-user social experience that way.
120
384030
3751
๊ทธ๋Ÿฐ ์‹์œผ๋กœ ์ƒํ˜ธ ์‚ฌ์šฉ์ž ์‚ฌํšŒ ๊ฒฝํ—˜์ด ๋ฉ๋‹ˆ๋‹ค.
06:27
And of course, a by-product of all of that is immensely rich virtual models
121
387805
4171
๋ฌผ๋ก , ๊ทธ ๋ชจ๋“  ๊ฒƒ์˜ ๋ถ€์‚ฐ๋ฌผ์€
์ง€๊ตฌ์˜ ๋ชจ๋“  ํฅ๋ฏธ๋กœ์šด ๋ถ€๋ถ„์˜
06:32
of every interesting part of the Earth,
122
392000
1968
์—„์ฒญ๋‚˜๊ฒŒ ํ’์š”๋กœ์šด ๋ฒ„์ถ”์–ผ ๋ชจ๋ธ๋“ค์ž…๋‹ˆ๋‹ค.
06:33
collected not just from overhead flights and from satellite images
123
393992
4487
๋จธ๋ฆฌ ์œ„๋ฅผ ์ง€๋‚˜๊ฐ€๋Š” ๋น„ํ–‰๊ธฐ๋‚˜ ์œ„์„ฑ ์ด๋ฏธ์ง€๋“ค์—์„œ ์ˆ˜ํ•ฉํ•œ ๊ฒƒ์ด ์•„๋‹ˆ๋ผ
06:38
and so on, but from the collective memory.
124
398503
2052
์ง‘๋‹จ์ ์ธ ๊ธฐ์–ต์—์„œ ์ˆ˜ํ•ฉํ•œ ๊ฒƒ์ž…๋‹ˆ๋‹ค.
06:40
Thank you so much.
125
400579
1094
๋Œ€๋‹จํžˆ ๊ฐ์‚ฌํ•ฉ๋‹ˆ๋‹ค.
06:41
(Applause)
126
401697
6863
(๋ฐ•์ˆ˜).
06:51
(Applause ends)
127
411967
1001
06:52
Chris Anderson: Do I understand this right?
128
412992
2326
ํฌ๋ฆฌ์Šค ์•ค๋”์Šจ: ์ œ๊ฐ€ ์ œ๋Œ€๋กœ ์ดํ•ด๋ฅผ ํ•˜๊ณ  ์žˆ๋‚˜์š”? ๋ง์”€ํ•˜์‹  ์†Œํ”„ํŠธ์›จ์–ด๊ฐ€
06:55
What your software is going to allow,
129
415342
2497
06:57
is that at some point, really within the next few years,
130
417863
3476
์–ด๋–ค ์‹œ์ ์—์„œ, ์ •๋ง ์•ž์œผ๋กœ ๋ช‡ ๋…„ ๋‚ด์—,
07:01
all the pictures that are shared by anyone across the world
131
421363
4235
์ „์„ธ๊ณ„์—์„œ ๋ˆ„๊ตฌ๋‚˜ ๊ณต์œ ํ•˜๋Š” ๋ชจ๋“  ์‚ฌ์ง„๋“ค์ด
07:05
are going to link together?
132
425622
1561
๊ธฐ๋ณธ์ ์œผ๋กœ ์„œ๋กœ ์—ฐ๊ฒฐ๋˜๋„๋ก ํ•ด์ค„ ๊ฒƒ์ด๋ผ๋Š” ๊ฑด๊ฐ€์š”?
07:07
BAA: Yes. What this is really doing is discovering,
133
427207
2387
BAA: ์˜ˆ. ์ด๊ฒƒ์ด ์‹ค์ œ๋กœ ํ•˜๊ณ  ์žˆ๋Š” ๊ฒƒ์€ ๋ฐœ๊ฒฌํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
07:09
creating hyperlinks, if you will, between images.
134
429618
2358
๊ทธ๋ฆฌ๊ณ  ์ด๋ฏธ์ง€๋“ค ๊ฐ„์— ํ•˜์ดํผ๋งํฌ๋ฅผ ๋งŒ๋“œ๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
07:12
It's doing that based on the content inside the images.
135
432000
2584
์ด๋ฏธ์ง€ ๋‚ด๋ถ€์— ์žˆ๋Š” ์ปจํ…์ธ ๋ฅผ ๊ธฐ๋ฐ˜์œผ๋กœ
๊ทธ๋ ‡๊ฒŒ ํ•˜๋Š” ๊ฒƒ์ž…๋‹ˆ๋‹ค.
07:14
And that gets really exciting when you think about the richness
136
434608
3022
๊ทธ ์ˆ˜๋งŽ์€ ์ด๋ฏธ์ง€๋“ค์ด ๊ฐ€์ง€๊ณ  ์žˆ๋Š” ์˜๋ฏธ๋ก ์ ์ธ ์ •๋ณด์˜
07:17
of the semantic information a lot of images have.
137
437654
2304
ํ’์š”๋กœ์›€์— ๋Œ€ํ•ด ์ƒ๊ฐํ•ด ๋ณด์‹œ๋ฉด ์ •๋ง ํฅ๋ถ„ํ•˜๊ฒŒ ๋ฉ๋‹ˆ๋‹ค.
07:19
Like when you do a web search for images,
138
439982
1960
์ด๋ฏธ์ง€๋ฅผ ์ฐพ์•„ ์›น๊ฒ€์ƒ‰์„ ํ•  ๋•Œ์ฒ˜๋Ÿผ,
07:21
you type in phrases,
139
441966
1245
์›ํ•˜๋Š” ๊ตฌ์ ˆ์„ ์ž…๋ ฅํ•ฉ๋‹ˆ๋‹ค. ์›นํŽ˜์ด์ง€์˜ ํ…์ŠคํŠธ๊ฐ€
07:23
and the text on the web page is carrying a lot of information
140
443235
2900
๊ทธ ์‚ฌ์ง„์ด ๋ฌด์—‡์ธ์ง€์— ๋Œ€ํ•ด ์ˆ˜๋งŽ์€ ์ •๋ณด๋ฅผ ๋‹ด๊ณ  ์žˆ์Šต๋‹ˆ๋‹ค.
07:26
about what that picture is of.
141
446159
1502
07:27
What if that picture links to all of your pictures?
142
447685
2391
์ด์ œ, ๊ทธ ์‚ฌ์ง„์ด ์—ฌ๋Ÿฌ๋ถ„์˜ ์‚ฌ์ง„ ๋ชจ๋‘์™€ ์—ฐ๊ฒฐ๋˜๋ฉด ์–ด๋–ป๊ฒŒ ๋ ๊นŒ์š”?
๊ทธ๋Ÿฌ๋ฉด, ์˜๋ฏธ๋ก ์ ์ธ ์ƒํ˜ธ ์—ฐ๊ฒฐ์˜ ์–‘๊ณผ
07:30
The amount of semantic interconnection and richness
143
450100
2413
๊ทธ๋กœ๋ถ€ํ„ฐ ๋‚˜์˜ค๋Š” ํ’์š”๋กœ์›€์˜ ์–‘์ด
07:32
that comes out of that is really huge.
144
452537
1854
์ •๋ง๋กœ ๋ง‰๋Œ€ํ•ฉ๋‹ˆ๋‹ค. ๊ณ ์ „์ ์ธ ๋„คํŠธ์›Œํฌ ํšจ๊ณผ์ž…๋‹ˆ๋‹ค.
07:34
It's a classic network effect.
145
454415
1449
07:35
CA: Truly incredible. Congratulations.
146
455888
2024
ํฌ๋ฆฌ์Šค ์•ค๋”์Šจ: ๋ธ”๋ ˆ์ฆˆ, ์ •๋ง ๋†€๋ž์Šต๋‹ˆ๋‹ค. ์ถ•ํ•˜ํ•ฉ๋‹ˆ๋‹ค.
BAA: ๋Œ€๋‹จํžˆ ๊ฐ์‚ฌํ•ฉ๋‹ˆ๋‹ค.
์ด ์›น์‚ฌ์ดํŠธ ์ •๋ณด

์ด ์‚ฌ์ดํŠธ๋Š” ์˜์–ด ํ•™์Šต์— ์œ ์šฉํ•œ YouTube ๋™์˜์ƒ์„ ์†Œ๊ฐœํ•ฉ๋‹ˆ๋‹ค. ์ „ ์„ธ๊ณ„ ์ตœ๊ณ ์˜ ์„ ์ƒ๋‹˜๋“ค์ด ๊ฐ€๋ฅด์น˜๋Š” ์˜์–ด ์ˆ˜์—…์„ ๋ณด๊ฒŒ ๋  ๊ฒƒ์ž…๋‹ˆ๋‹ค. ๊ฐ ๋™์˜์ƒ ํŽ˜์ด์ง€์— ํ‘œ์‹œ๋˜๋Š” ์˜์–ด ์ž๋ง‰์„ ๋”๋ธ” ํด๋ฆญํ•˜๋ฉด ๊ทธ๊ณณ์—์„œ ๋™์˜์ƒ์ด ์žฌ์ƒ๋ฉ๋‹ˆ๋‹ค. ๋น„๋””์˜ค ์žฌ์ƒ์— ๋งž์ถฐ ์ž๋ง‰์ด ์Šคํฌ๋กค๋ฉ๋‹ˆ๋‹ค. ์˜๊ฒฌ์ด๋‚˜ ์š”์ฒญ์ด ์žˆ๋Š” ๊ฒฝ์šฐ ์ด ๋ฌธ์˜ ์–‘์‹์„ ์‚ฌ์šฉํ•˜์—ฌ ๋ฌธ์˜ํ•˜์‹ญ์‹œ์˜ค.

https://forms.gle/WvT1wiN1qDtmnspy7