Blaise Aguera y Arcas: Jaw-dropping Photosynth demo

46,176 views ใƒป 2007-06-26

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

00:25
What I'm going to show you first, as quickly as I can,
0
25000
2548
ืžื” ืฉืื ื™ ื”ื•ืœืš ืœื”ืจืื•ืช ืœื›ื ืงื•ื“ื ื›ื•ืœ, ืžื”ืจ ื›ื›ื•ืœ ื”ืืคืฉืจ,
00:27
is some foundational work, some new technology
1
27572
3769
ื”ื•ื ืขื‘ื•ื“ื” ื‘ืกื™ืกื™ืช, ื˜ื›ื ื•ืœื•ื’ื™ื” ื—ื“ืฉื”
00:31
that we brought to Microsoft as part of an acquisition
2
31365
2611
ืฉื”ื‘ืื ื• ืœืžื™ืงืจื•ืกื•ืคื˜ ื‘ืžืกื’ืจืช ืจื›ื™ืฉื”
00:34
almost exactly a year ago.
3
34000
1821
ืœืคื ื™ ืฉื ื” ื›ืžืขื˜. ื–ื• ืกื™ื“ืจื’ื•ืŸ.
00:35
This is Seadragon, and it's an environment
4
35845
2368
ืกื‘ื™ื‘ื” ืฉื‘ื” ื ื™ืชืŸ ื‘ืื•ืคืŸ ืžืงื•ืžื™ ืื• ืžืจื—ื•ืง
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
ืœืคืขื•ืœ ืขื ื›ืžื•ื™ื•ืช ืขืฆื•ืžื•ืช ืฉืœ ื ืชื•ื ื™ื ื—ื–ื•ืชื™ื™ื.
00:43
We're looking at many, many gigabytes of digital photos here
7
43165
3404
ืื ื• ืžื‘ื™ื˜ื™ื ื‘ื”ืžื•ื ื™ ื’'ื™ื’ื”-ื‘ื™ื™ื˜ ืฉืœ ืชืžื•ื ื•ืช ื“ื™ื’ื™ื˜ืœื™ื•ืช
00:46
and kind of seamlessly and continuously zooming in,
8
46593
2915
ื•ืžื’ื“ื™ืœื™ื ืืช ื”ืชืฆื•ื’ื” ื‘ืื•ืคืŸ ื—ืœืง ื•ืžืชืžืฉืš,
00:49
panning through it, rearranging it in any way we want.
9
49532
2545
ื•ื ืขื™ื ื‘ืชื•ืš ื”ืชืžื•ื ื•ืช, ืžืกื“ืจื™ื ืื•ืชืŸ ืื™ืš ืฉื ืจืฆื”,
00:52
And it doesn't matter how much information we're looking at,
10
52389
3587
ื•ื–ื” ืœื ืžืฉื ื” ื‘ื›ืžื” ืžื™ื“ืข ืื ื• ืžื‘ื™ื˜ื™ื,
00:56
how big these collections are or how big the images are.
11
56000
2976
ื›ืžื” ื’ื“ื•ืœื™ื ื”ืื•ืกืคื™ื ืื• ื›ืžื” ื’ื“ื•ืœื•ืช ื”ืชืžื•ื ื•ืช.
00:59
Most of them are ordinary digital camera photos,
12
59000
2286
ืจื•ื‘ืŸ ืชืžื•ื ื•ืช ืžืžืฆืœืžื•ืช ื“ื™ื’ื™ื˜ืœื™ื•ืช ืคืฉื•ื˜ื•ืช,
01:01
but this one, for example, is a scan from the Library of Congress,
13
61310
3144
ืืš ื–ื•, ืœืžืฉืœ, ืกืจื™ืงื” ืžืกืคืจื™ื™ืช ื”ืงื•ื ื’ืจืก,
01:04
and it's in the 300 megapixel range.
14
64478
2818
ื•ื–ื” ื˜ื•ื•ื— ืฉืœ 300 ืžื’ื”-ืคื™ืงืกืœ.
01:07
It doesn't make any difference
15
67320
1656
ื–ื” ืœื ืžืฉื ื”,
01:09
because the only thing that ought to limit the performance of a system like this one
16
69000
4144
ื›ื™ ื”ื“ื‘ืจ ื”ื™ื—ื™ื“ ืฉื™ื›ื•ืœ ืœื”ื’ื‘ื™ืœ ื‘ื™ืฆื•ืขื™ื ืฉืœ
ืžืขืจื›ืช ื›ื–ืืช ื”ื•ื ืžืกืคืจ ื”ืคื™ืงืกืœื™ื ื‘ืžืกืš
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 ืจื’ืข ื ืชื•ืŸ. ื–ื• ื’ื ืืจื›ื™ื˜ืงื˜ื•ืจื” ื’ืžื™ืฉื” ืžืื•ื“.
01:17
This is an entire book, so this is an example of non-image data.
19
77963
3727
ื–ื” ืกืคืจ ืฉืœื, ื“ื•ื’ืžื” ืœื ืชื•ื ื™ื ืœื-ืชืžื•ื ืชื™ื™ื.
01:21
This is "Bleak House" by Dickens.
20
81714
2787
ื–ื” "ื”ื‘ื™ืช ื”ืขื’ื•ื" ืžืืช ื“ื™ืงื ืก. ื›ืœ ื˜ื•ืจ ื”ื•ื ืคืจืง.
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
ื›ื“ื™ ืœื”ื•ื›ื™ื— ืฉื–ื” ื‘ืืžืช ื˜ืงืกื˜ ื•ืœื ืชืžื•ื ื”,
01:31
we can do something like so, to really show
23
91000
2048
ืืคืฉืจ ืœืขืฉื•ืช ืžืฉื”ื• ื›ื–ื”, ื›ื“ื™ ืœื”ืจืื•ืช ื‘ืืžืช
01:33
that this is a real representation of the text; it's not a picture.
24
93072
3192
ืฉื–ื• ืชืฆื•ื’ื” ืืžื™ืชื™ืช ืฉืœ ื”ื˜ืงืกื˜; ื–ื• ืœื ืชืžื•ื ื”.
01:36
Maybe this is an artificial way to read an e-book.
25
96288
2664
ืื•ืœื™ ื–ื• ื“ืจืš ืžืœืื›ื•ืชื™ืช ืœืงืจื•ื ืกืคืจ ืืœืงื˜ืจื•ื ื™.
01:38
I wouldn't recommend it.
26
98976
1200
ืœื ื”ื™ื™ืชื™ ืžืžืœื™ืฅ ืขืœ ื›ืš.
01:40
This is a more realistic case, an issue of The Guardian.
27
100200
2848
ื”ื ื” ืžืงืจื” ื™ื•ืชืจ ืžืฆื™ืื•ืชื™. ื–ื” ื’ื™ืœื™ื•ืŸ ืฉืœ ื”"ื’ืืจื“ื™ืืŸ".
01:43
Every large image is the beginning of a section.
28
103072
2286
ื›ืœ ืชืžื•ื ื” ื’ื“ื•ืœื” ืžื”ื•ื•ื” ืชื—ื™ืœืช ื”ืงื˜ืข.
01:45
And this really gives you the joy and the good experience
29
105382
2904
ื•ื–ื” ื‘ืืžืช ืžืขื ื™ืง ืืช ื”ื”ื ืื” ื•ื”ื—ื•ื•ื™ื” ื”ื ืขื™ืžื”
01:48
of reading the real paper version of a magazine or a newspaper,
30
108310
5183
ืฉืœ ืงืจื™ืืช ื’ื™ืœื™ื•ืŸ ื ื™ื™ืจ ืืžื™ืชื™ ืฉืœ ืžื’ื–ื™ืŸ ืื• ืขื™ืชื•ืŸ,
01:53
which is an inherently multi-scale kind of medium.
31
113517
2435
ืฉื”ื•ื ื‘ืžื”ื•ืชื• ืžื“ื™ื•ื ืขื ืงื ื™ ืžื™ื“ื” ืžืจื•ื‘ื™ื.
01:55
We've done something
32
115976
1000
ืขืฉื™ื ื• ื’ื ืžืฉื”ื• ืงื˜ืŸ
01:57
with the corner of this particular issue of The Guardian.
33
117000
2976
ืขื ืคื™ื ื” ืงื˜ื ื” ืฉืœ ื’ื™ืœื™ื•ืŸ ืžืกื•ื™ื ืฉืœ ื”"ื’ืืจื“ื™ืืŸ".
02:00
We've made up a fake ad that's very high resolution --
34
120000
2976
ื”ื›ื ื• ืคืจืกื•ืžืช ืžื–ื•ื™ืคืช ื‘ืจื–ื•ืœื•ืฆื™ื” ื’ื‘ื•ื”ื” ืžืื•ื“,
02:03
much higher than in an ordinary ad --
35
123000
2198
ื’ื‘ื•ื”ื” ื‘ื”ืจื‘ื” ืžืฉืชื•ื›ืœื• ืœื›ืœื•ืœ ื‘ืคืจืกื•ืžืช ืจื’ื™ืœื”,
02:05
and we've embedded extra content.
36
125222
1754
ื•ื”ื˜ืžืขื ื• ืชื•ื›ืŸ ื ื•ืกืฃ.
02:07
If you want to see the features of this car, you can see it here.
37
127000
3048
ืื ืชืจืฆื• ืœืจืื•ืช ืืช ื”ืชื›ื•ื ื•ืช ืฉืœ ื”ืžื›ื•ื ื™ืช, ืชื•ื›ืœื• ืœืจืื•ืช ื›ืืŸ.
02:10
Or other models, or even technical specifications.
38
130072
4180
ืื• ื“ื’ืžื™ื ืื—ืจื™ื, ืื• ืืคื™ืœื• ืžืคืจื˜ื™ื ื˜ื›ื ื™ื™ื.
02:14
And this really gets at some of these ideas
39
134276
3315
ื•ื–ื” ื‘ืืžืช ื ื•ื’ืข ื‘ื›ืžื” ืžื”ืจืขื™ื•ื ื•ืช
02:17
about really doing away with those limits on screen real estate.
40
137615
4661
ืœื”ื™ืคื˜ืจ ืžืžื’ื‘ืœื•ืช ื”ื ื“ืœ"ืŸ ืฉืœ ื”ืžืกืš.
02:22
We hope that this means no more pop-ups
41
142300
2111
ืื ื• ืžืงื•ื•ื™ื ืฉืคื™ืจื•ืฉ ื”ื“ื‘ืจ ื”ื•ื ืกื•ืฃ ืœื—ืœื•ื ื•ืช ื”ืžื•ืงืคืฆื™ื
02:24
and other rubbish like that -- shouldn't be necessary.
42
144435
2541
ื•ื–ื‘ืœ ืื—ืจ. ืื™ืŸ ื‘ื”ื ืฆื•ืจืš.
02:27
Of course, mapping is one of those obvious applications
43
147000
2658
ื›ืžื•ื‘ืŸ, ืžื™ืคื•ื™ ื”ื•ื ืื—ื“ ื”ื™ื™ืฉื•ืžื™ื ื”ืžืชื‘ืงืฉื™ื
02:29
for a technology like this.
44
149682
1294
ืฉืœ ื˜ื›ื ื•ืœื•ื’ื™ื” ื›ื–ืืช.
02:31
And this one I really won't spend any time on,
45
151000
2191
ื•ืื ื™ ืœื ืจื•ืฆื” ืœื‘ื–ื‘ื– ืขืœ ื–ื” ื–ืžืŸ,
02:33
except to say that we have things to contribute to this field as well.
46
153215
3334
ืจืง ืื•ืžืจ ืฉื™ืฉ ืœื ื• ืžื” ืœืชืจื•ื ื’ื ื‘ืชื—ื•ื ื”ื–ื”.
02:37
But those are all the roads in the U.S.
47
157213
1858
ื”ื ื” ื›ืœ ื”ื›ื‘ื™ืฉื™ื ื‘ืืจืฆื•ืช ื”ื‘ืจื™ืช
02:39
superimposed on top of a NASA geospatial image.
48
159095
4565
ืžื•ืฆื’ื™ื ืขืœ ืชืžื•ื ื” ื’ื™ืื•-ืžืจื—ื‘ื™ืช ืฉืœ ื ืืกื.
02:44
So let's pull up, now, something else.
49
164000
1976
ืื– ื‘ื•ืื• ื ื‘ื™ื ืžืฉื”ื• ืื—ืจ.
02:46
This is actually live on the Web now; you can go check it out.
50
166000
2976
ื–ื” ื›ืจื’ืข ื™ืฉืจ ืžื”ืื™ื ื˜ืจื ื˜; ืืชื ื™ื›ื•ืœื™ื ืœืœื›ืช ืœืฉื ื•ืœื‘ื“ื•ืง.
02:49
This is a project called Photosynth, which marries two different technologies.
51
169000
3704
ื–ื”ื• ืคืจื•ื™ืงื˜ ื‘ืฉื ืคื•ื˜ื•ืกื™ื ืช,
ืฉืžืฉื“ืš ื‘ื™ืŸ ืฉืชื™ ื˜ื›ื ื•ืœื•ื’ื™ื•ืช ืฉื•ื ื•ืช.
02:52
One of them is Seadragon
52
172728
1248
ืื—ืช ืžื”ื ื”ื™ื ืกื™ื“ืจื’ื•ืŸ
02:54
and the other is some very beautiful computer-vision research
53
174000
2906
ื•ื”ืฉื ื™ื™ื” ื”ื™ื ืžื—ืงืจ ื—ื–ื•ืชื™ ืžืžื•ื—ืฉื‘ ื™ืคื”
02:56
done by Noah Snavely, a graduate student at the University of Washington,
54
176930
3462
ืฉืœ ื ื•ื— ืฉื ื™ื™ื‘ืœื™, ืกื˜ื•ื“ื ื˜ ื‘ืื•ื ื™ื‘ืจืกื™ื˜ืช ื•ื•ืฉื™ื ื’ื˜ื•ืŸ,
03:00
co-advised by Steve Seitz at U.W.
55
180416
1829
ื‘ื™ื™ืขื•ืฅ ืžืฉื•ืชืฃ ืฉืœ ืกื˜ื™ื‘ ืกื™ื™ืฅ ืžื”ืื•ื ื™ื‘ืจืกื™ื˜ื”
03:02
and Rick Szeliski at Microsoft Research.
56
182269
1978
ื•ืจื™ืง ืกืœื™ืกืงื™ ืžืžื—ืงืจ ืžื™ืงืจื•ืกื•ืคื˜. ืฉื™ืชื•ืฃ ืคืขื•ืœื” ืžืื•ื“ ื™ืคื”.
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
ืื– ื–ื” ื™ืฉื™ืจ ืžื”ืจืฉืช. ื–ื” ืžื•ืคืขืœ ื‘ืืžืฆืขื•ืช ืกื™ื“ืจื’ื•ืŸ.
03:09
You can see that when we do these sorts of views,
59
189544
2504
ืืคืฉืจ ืœืจืื•ืช ืฉื›ืฉืื ื—ื ื• ืขื•ืฉื™ื ืชืฆื•ื’ื•ืช ื›ืืœื•,
03:12
where we can dive through images
60
192072
1723
ืฉื‘ื”ืŸ ืื ื—ื ื• ืฆื•ืœืœื™ื ืœืชื•ืš ื”ืชืžื•ื ื•ืช
03:13
and have this kind of multi-resolution experience.
61
193819
2334
ื•ืžืงื‘ืœื™ื ื—ื•ื•ื™ื” ื›ื–ืืช ืขื ืจื™ื‘ื•ื™ ืจื–ื•ืœื•ืฆื™ื•ืช.
03:16
But the spatial arrangement of the images here is actually meaningful.
62
196177
3799
ืื‘ืœ ืœืกื™ื“ื•ืจ ื”ืžืจื—ื‘ื™ ืฉืœ ื”ืชืžื•ื ื•ืช ื›ืืŸ ื™ืฉ ืžืฉืžืขื•ืช.
03:20
The computer vision algorithms have registered these images together
63
200000
3191
ืืœื’ื•ืจื™ืชืžื™ ื”ืชืฆื•ื’ื” ืฉืœ ื”ืžื—ืฉื‘ ืจืฉืžื• ืืช ื”ืชืžื•ื ื•ืช ื‘ื™ื—ื“,
03:23
so that they correspond to the real space in which these shots --
64
203215
3761
ื›ืš ืฉื”ืŸ ื™ืชืืžื• ืœื—ืœืœ ื”ืืžื™ืชื™ ืฉื‘ื• ื”ืฆื™ืœื•ืžื™ื
03:27
all taken near Grassi Lakes in the Canadian Rockies --
65
207000
3300
ื ืœืงื—ื•, ืœื™ื“ ื’ืจืืกื™ ืœื™ื™ืงืก ื‘ื”ืจื™ ื”ืจื•ืงื™ ื‘ืงื ื“ื”,
03:30
all these shots were taken.
66
210324
1663
ืื– ืืคืฉืจ ืœืจืื•ืช ื›ืืŸ ืจื›ื™ื‘ื™ื
03:32
So you see elements here
67
212011
1467
03:33
of stabilized slide-show or panoramic imaging,
68
213502
6013
ืฉืœ ืชืฆื•ื’ืช ืฉืงื•ืคื™ื•ืช ืžื™ื•ืฆื‘ืช ืื• ื”ื“ืžื™ื” ืคื ื•ืจืžื™ืช,
03:39
and these things have all been related spatially.
69
219539
2437
ื•ื‘ื™ืŸ ื›ืœ ื”ื“ื‘ืจื™ื ื”ืืœื” ื™ืฉ ืงืฉืจ ืžืจื—ื‘ื™.
03:42
I'm not sure if I have time to show you any other environments.
70
222000
3000
ืื ื™ ืœื ื‘ื˜ื•ื— ืื ื™ืฉ ืœื™ ื–ืžืŸ ืœื”ืจืื•ืช ืœื›ื ืกื‘ื™ื‘ื•ืช ืื—ืจื•ืช.
03:45
Some are much more spatial.
71
225024
1431
ื™ืฉ ื›ืžื” ื”ืจื‘ื” ื™ื•ืชืจ ืžืจื—ื‘ื™ื•ืช.
03:46
I would like to jump straight to one of Noah's original data-sets --
72
226479
3945
ืื ื™ ืขื•ื‘ืจ ื™ืฉืจ ืœืื—ื“ ืžืื•ืกืคื™ ื”ื ืชื•ื ื™ื ื”ืžืงื•ืจื™ื™ื ืฉืœ ื ื•ื—,
03:50
this is from an early prototype that we first got working this summer --
73
230448
3552
ื•ื–ื” ืžืื‘-ื˜ื™ืคื•ืก ืžื•ืงื“ื ืฉืœ ืคื•ื˜ื•ืกื™ื ืช
ืฉื”ืฆืœื—ื ื• ืœื”ืคืขื™ืœ ื‘ืงื™ืฅ,
03:54
to show you what I think
74
234024
1894
ื›ื“ื™ ืœื”ืจืื•ืช ืœื›ื ืืช
03:55
is really the punch line behind the Photosynth technology,
75
235942
3838
ืžื” ืฉืื ื™ ืจื•ืื” ื›ืฉื•ืจืช ื”ืžื—ืฅ ืฉืœ ื”ื˜ื›ื ื•ืœื•ื’ื™ื”,
03:59
It's not necessarily so apparent
76
239804
1561
ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ืฉืœ ืคื•ื˜ื•ืกื™ื ืช. ื•ืื™ืŸ ื–ื” ื‘ืจื•ืจ ื‘ื”ื›ืจื—
04:01
from looking at the environments we've put up on the website.
77
241389
2895
ืžื”ืกืชื›ืœื•ืช ื‘ืกื‘ื™ื‘ื•ืช ืฉื”ืขืœื™ื ื• ืœืืชืจ ื”ืื™ื ื˜ืจื ื˜.
04:04
We had to worry about the lawyers and so on.
78
244308
2177
ื”ื™ื™ื ื• ืฆืจื™ื›ื™ื ืœื”ื™ื–ื”ืจ ืžืขื•ืจื›ื™ ื“ื™ืŸ ื•ื›ืŸ ื”ืœืื”.
04:06
This is a reconstruction of Notre Dame Cathedral
79
246509
2301
ื–ื” ืฉื—ื–ื•ืจ ืฉืœ ื›ื ืกื™ื™ืช ื ื•ื˜ืจ ื“ืื,
04:08
that was done entirely computationally from images scraped from Flickr.
80
248834
3457
ื›ื•ืœื• ื‘ืืžืฆืขื•ืช ืžื—ืฉื‘,
ืžืชืžื•ื ื•ืช ืฉื ืืกืคื• ืž"ืคืœื™ืงืจ". ืคืฉื•ื˜ ืžืงืœื™ื“ื™ื ื ื•ื˜ืจ ื“ืื "ื‘"ืคืœื™ืงืจ",
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
ื•ืžืงื‘ืœื™ื ืชืžื•ื ื•ืช ืฉืœ ืื ืฉื™ื ื‘ื—ื•ืœืฆื•ืช ื˜ื™, ื•ืฉืœ ื”ืงืžืคื•ืก,
ื•ื›ืŸ ื”ืœืื”. ื•ื›ืœ ืื—ื“ ืžื”ืงื•ื ื•ืกื™ื ื”ื›ืชื•ืžื™ื ืžื™ื™ืฆื’ ืชืžื•ื ื”
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
ืฉื ืžืฆืื” ื›ืฉื™ื™ื›ืช ืœืžื•ื“ืœ ื”ื–ื”.
04:26
And so these are all Flickr images,
85
266000
1976
ืื– ื›ืœ ืืœื” ืชืžื•ื ื•ืช ืž"ืคืœื™ืงืจ",
04:28
and they've all been related spatially in this way.
86
268000
2976
ื•ื ื•ืฆืจื• ื‘ื™ื ื™ื”ืŸ ืงืฉืจื™ื ืžืจื—ื‘ื™ื™ื ื›ื›ื”.
04:31
We can just navigate in this very simple way.
87
271000
2334
ื•ื ื™ืชืŸ ืœื ื•ื•ื˜ ื‘ื“ืจืš ื”ืคืฉื•ื˜ื” ื”ื–ืืช.
04:35
(Applause)
88
275000
3920
(ื›ืคื™ื™ื)
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
ืืฃ ืคืขื ืœื ื—ืฉื‘ืชื™ ืฉื‘ืกื•ืฃ ืืขื‘ื•ื“ ื‘ืžื™ืงืจื•ืกื•ืคื˜.
04:46
It's very gratifying to have this kind of reception here.
91
286573
3000
ืžืฉืžื— ืœื–ื›ื•ืช ื‘ื›ื–ืืช ืงื‘ืœื” ื›ืืŸ.
04:49
(Laughter)
92
289597
3379
(ืฆื—ื•ืง)
04:53
I guess you can see this is lots of different types of cameras:
93
293000
5048
ืืชื ื‘ื˜ื— ื™ื›ื•ืœื™ื ืœืจืื•ืช
ืฉืืœื• ืกื•ื’ื™ื ืฉื•ื ื™ื ืฉืœ ืžืฆืœืžื•ืช:
04:58
it's everything from cell-phone cameras to professional SLRs,
94
298072
3161
ื›ืœ ื“ื‘ืจ ืžืžืฆืœืžื•ืช ืกืœื•ืœืจื™ื•ืช ืœืžืฆืœืžื•ืช ืจืคืœืงืก,
05:01
quite a large number of them, stitched together in this environment.
95
301257
3191
ื“ื™ ื”ืจื‘ื” ืžื”ืŸ, ืฉื ืชืคืจื•
ื‘ื™ื—ื“ ื‘ืกื‘ื™ื‘ื” ื”ื–ืืช.
05:04
If I can find some of the sort of weird ones --
96
304472
2632
ื•ืื ืื•ื›ืœ ืืžืฆื ื›ืžื” ืžื”ืžื•ื–ืจื™ื.
05:08
So many of them are occluded by faces, and so on.
97
308000
3322
ื”ืจื‘ื” ืžื”ืŸ ืžื•ืกืชืจื•ืช ืขืœ-ื™ื“ื™ ืคื ื™ื ื•ื›ืŸ ื”ืœืื”.
05:12
Somewhere in here there is actually a series of photographs -- here we go.
98
312595
4277
ืื™ืคืฉื”ื• ื›ืืŸ ื™ืฉ
ืกื“ืจืช ืฆื™ืœื•ืžื™ื, ื”ื ื” ื–ื”.
05:16
This is actually a poster of Notre Dame that registered correctly.
99
316896
3301
ื–ื• ื›ืจื–ื” ืฉืœ ื ื•ื˜ืจ ื“ืื ืฉื ืจืฉืžื” ื ื›ื•ืŸ.
05:20
We can dive in from the poster
100
320221
3216
ืืคืฉืจ ืœืฆืœื•ืœ ืžื”ื›ืจื–ื”
05:23
to a physical view of this environment.
101
323461
3810
ืœืชืฆื•ื’ื” ืคื™ื–ื™ืช ืฉืœ ื”ืกื‘ื™ื‘ื”.
05:31
What the point here really is
102
331421
1866
ื”ืจืขื™ื•ืŸ ื›ืืŸ ื”ื•ื ืฉืืคืฉืจ ืœืขืฉื•ืช ื“ื‘ืจื™ื
05:33
is that we can do things with the social environment.
103
333311
2591
ืขื ื”ืกื‘ื™ื‘ื” ื”ื—ื‘ืจืชื™ืช. ื–ื” ืœื•ืงื— ื ืชื•ื ื™ื ืžื›ื•ืœื,
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
ืžื›ืœ ื”ื–ื™ื›ืจื•ืŸ ื”ืงื•ืœืงื˜ื™ื‘ื™
ืฉืœ ื”ืื•ืคืŸ ืฉื‘ื• ื”ืขื•ืœื ื ืจืื”,
05:42
and link all of that together.
106
342847
1749
ื•ืžืงืฉืจ ื”ื›ื•ืœ ื‘ื™ื—ื“.
05:44
Those photos become linked, and they make something emergent
107
344620
2839
ื›ืœ ื”ืชืžื•ื ื•ืช ื”ืืœื” ืžืงื•ืฉืจื•ืช ื™ื—ื“,
ื•ืžืฆื™ื’ื•ืช ืžืฉื”ื•
05:47
that's greater than the sum of the parts.
108
347483
1953
ืฉื’ื“ื•ืœ ื™ื•ืชืจ ืžืกื›ื•ื ื—ืœืงื™ื•.
05:49
You have a model that emerges of the entire Earth.
109
349460
2356
ืžืชื’ื‘ืฉ ืžื•ื“ืœ ืฉืœ ื”ืขื•ืœื ื›ื•ืœื•.
05:51
Think of this as the long tail to Stephen Lawler's Virtual Earth work.
110
351840
4077
ื—ืฉื‘ื• ืขืœ ื›ืš ื›ืขืœ ื–ื ื‘ื• ื”ืืจื•ืš ืฉืœ "ื•ื™ืจืฆ'ื•ืืœ ืืจืช'" ืฉืœ ืกื˜ื™ื‘ืŸ ืœื•ืœืจ.
05:55
And this is something that grows in complexity as people use it,
111
355941
3200
ื•ื–ื” ื“ื‘ืจ ืฉื ืขืฉื” ื™ื•ืชืจ ื•ื™ื•ืชืจ ืžื•ืจื›ื‘
ื›ื›ื•ืœ ืฉืื ืฉื™ื ืžืฉืชืžืฉื™ื ื‘ื•, ื•ืืฉืจ ื”ืชื•ืขืœืช ืžืžื ื• ื’ื“ืœื”
05:59
and whose benefits become greater to the users as they use it.
112
359165
3811
ืœืžืฉืชืžืฉื™ื ื›ื›ื•ืœ ืฉื”ื ืžืฉืชืžืฉื™ื ื‘ื•.
06:03
Their own photos are getting tagged with meta-data that somebody else entered.
113
363000
3692
ื”ืชืžื•ื ื•ืช ืฉืœื”ืŸ ืžืชื•ื™ื’ื•ืช ื‘ืžื˜ื”-ื ืชื•ื ื™ื
ืฉืžื™ืฉื”ื• ืื—ืจ ืžื–ื™ืŸ.
06:06
If somebody bothered to tag all of these saints
114
366716
3360
ืื ืžื™ืฉื”ื• ืžืชื™ื™ื’ ืืช ื›ืœ ื”ืงื“ื•ืฉื™ื ื”ืืœื”
06:10
and say who they all are, then my photo of Notre Dame Cathedral
115
370100
2953
ื•ืžืกืคืจ ืžื™ ื”ื, ืื– ื”ืชืžื•ื ื” ืฉืœื™ ืžื ื•ื˜ืจ ื“ืื
06:13
suddenly gets enriched with all of that data,
116
373077
2098
ืžื•ืขืฉืจืช ื‘ื›ืœ ื”ื ืชื•ื ื™ื ื”ืœืœื•,
06:15
and I can use it as an entry point to dive into that space,
117
375199
2777
ื•ื ื™ืชืŸ ืœื”ืฉืชืžืฉ ื‘ื”ื ื›ื ืงื•ื“ืช ื›ื ื™ืกื” ืœื—ืœืœ ื”ื–ื”,
06:18
into that meta-verse, using everybody else's photos,
118
378000
2681
ืœืขื•ืœื ื”ื•ื™ืจื˜ื•ืืœื™ ื”ื ืชื•ืŸ, ื‘ืืžืฆืขื•ืช ื”ืชืžื•ื ื•ืช ืฉืœ ื›ืœ ื”ืื—ืจื™ื,
06:20
and do a kind of a cross-modal
119
380705
3301
ื•ืœื—ื•ื•ืช ืžืขื™ืŸ ื—ื•ื•ื™ื” ื—ื‘ืจืชื™ืช
06:24
and cross-user social experience that way.
120
384030
3751
ื‘ื™ืŸ ืžื•ื“ืœื™ื ื•ื‘ื™ืŸ ืžืฉืชืžืฉื™ื.
06:27
And of course, a by-product of all of that is immensely rich virtual models
121
387805
4171
ื•ื›ืžื•ื‘ืŸ, ื™ืฉื ื• ืชื•ืฆืจ ื”ืœื•ื•ืื™
ืฉืœ ืžื•ื“ืœื™ื ื•ื™ืจื˜ื•ืืœื™ื™ื ืขืฉื™ืจื™ื ื‘ื™ื•ืชืจ
06:32
of every interesting part of the Earth,
122
392000
1968
ืฉืœ ื›ืœ ื—ืœืง ืžืขื ื™ื™ืŸ ื‘ืขื•ืœื, ืฉื ืืกืคื•
06:33
collected not just from overhead flights and from satellite images
123
393992
4487
ืœื ืจืง ืžืชืฆืœื•ืžื™ ืื•ื•ื™ืจ ื•ืœื•ื•ื™ืŸ
06:38
and so on, but from the collective memory.
124
398503
2052
ื•ื›ืŸ ื”ืœืื”, ืืœื ืžื”ื–ื™ื›ืจื•ืŸ ื”ืงื•ืœืงื˜ื™ื‘ื™.
06:40
Thank you so much.
125
400579
1094
ืชื•ื“ื” ืจื‘ื” ืœื›ื.
06:41
(Applause)
126
401697
6863
(ื›ืคื™ื™ื)
06:51
(Applause ends)
127
411967
1001
06:52
Chris Anderson: Do I understand this right?
128
412992
2326
ื›ืจื™ืก ืื ื“ืจืกื•ืŸ: ืื ื™ ืžื‘ื™ืŸ ื ื›ื•ืŸ? ืžื” ืฉื”ืชื•ื›ื ื” ืฉืœื›ื ืชืืคืฉืจ
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
ื”ื•ื ืฉื‘ืฉืœื‘ ืžืกื•ื™ื, ื‘ืžื”ืœืš ื”ืฉื ื™ื ื”ืงืจื•ื‘ื•ืช ืžืžืฉ,
07:01
all the pictures that are shared by anyone across the world
131
421363
4235
ื›ืœ ื”ืชืžื•ื ื•ืช ืฉืžืฉืชืคื™ื ื›ืœ ื”ืื ืฉื™ื ื‘ืขื•ืœื
07:05
are going to link together?
132
425622
1561
ื™ื”ื™ื• ืžืงื•ืฉืจื•ืช ื‘ื™ื—ื“?
07:07
BAA: Yes. What this is really doing is discovering,
133
427207
2387
ื‘ื"ื: ื›ืŸ. ืžื” ืฉื–ื” ื‘ืขืฆื ืขื•ืฉื” ื–ื” ื’ื™ืœื•ื™.
07:09
creating hyperlinks, if you will, between images.
134
429618
2358
ื–ื” ื™ื•ืฆืจ ื”ื™ืคืจ-ืงื™ืฉื•ืจื™ื, ืื ืชืจืฆื”, ื‘ื™ืŸ ืชืžื•ื ื•ืช.
07:12
It's doing that based on the content inside the images.
135
432000
2584
ื•ื–ื” ืขื•ืฉื” ื–ืืช
ื‘ื”ืชื‘ืกืก ืขืœ ืชื•ื›ืŸ ื‘ืชื•ืš ื”ืชืžื•ื ื•ืช.
07:14
And that gets really exciting when you think about the richness
136
434608
3022
ื•ื–ื” ืžืื•ื“ ืžืจื’ืฉ ืœื—ืฉื•ื‘ ืขืœ ื”ืขื•ืฉืจ
07:17
of the semantic information a lot of images have.
137
437654
2304
ืฉืœ ื”ืžื™ื“ืข ื”ืกืžื ื˜ื™ ืฉื ืžืฆื ื‘ื”ืจื‘ื” ืžื”ืชืžื•ื ื•ืช ื”ืืœื•.
07:19
Like when you do a web search for images,
138
439982
1960
ืœืžืฉืœ ื›ืฉืžื—ืคืฉื™ื ืชืžื•ื ื•ืช ื‘ืื™ื ื˜ืจื ื˜,
07:21
you type in phrases,
139
441966
1245
ืžืงืœื™ื“ื™ื ืฆื™ืจื•ืคื™ ืžื™ืœื™ื, ื•ื”ื˜ืงืกื˜ ื‘ื“ืฃ ื”ืื™ื ื˜ืจื ื˜
07:23
and the text on the web page is carrying a lot of information
140
443235
2900
ืžื›ื™ืœ ื”ืžื•ืŸ ืžื™ื“ืข ืื•ื“ื•ืช ื ื•ืฉื ื”ืชืžื•ื ื”.
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
ืžื” ืงื•ืจื” ืื ื”ืชืžื•ื ื” ืžืงื•ืฉืจืช ืœื›ืœ ื”ืชืžื•ื ื•ืช ืฉืœืš?
ืื– ื”ื›ืžื•ืช ืฉืœ ื”ืงื™ืฉื•ืจื™ื ื”ื”ื“ื“ื™ื™ื ื”ืกืžื ื˜ื™ื™ื
07:30
The amount of semantic interconnection and richness
143
450100
2413
ื•ื”ืขื•ืฉืจ ืฉืžื’ื™ืข ืžื›ืš
07:32
that comes out of that is really huge.
144
452537
1854
ืขืฆื•ืžื™ื ื‘ืืžืช. ื–ื”ื• ืืคืงื˜ ืจืฉืช ืงืœืืกื™.
07:34
It's a classic network effect.
145
454415
1449
07:35
CA: Truly incredible. Congratulations.
146
455888
2024
ื›"ื: ื‘ืœื™ื™ื–, ื–ื” ื‘ืืžืช ืžื“ื”ื™ื. ื‘ืจื›ื•ืชื™ื™.
.ื‘ื"ื: ืชื•ื“ื” ืจื‘ื”
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7