TED2007
Stephen Lawler: Tour Microsoft's Virtual Earth
Filmed:
Readability: 4.5
338,924 views
Microsoft's Stephen Lawler gives a whirlwind tour of Virtual Earth, moving up, down and through its hyper-real cityscapes with dazzlingly fluidity, a remarkable feat that requires staggering amounts of data to bring into focus.
Stephen Lawler - General manager of Microsoft's Virtual Earth
Stephen Lawler and the Virtual Earth team have created an addictively interactive 3D world that is poised to reinvent our view of advertising, gaming, weather/traffic reporting, instant messaging and more. Full bio
Stephen Lawler and the Virtual Earth team have created an addictively interactive 3D world that is poised to reinvent our view of advertising, gaming, weather/traffic reporting, instant messaging and more. Full bio
Double-click the English transcript below to play the video.
00:25
What I want to talk to you about today is
0
0
3000
00:28
virtual worlds, digital globes, the 3-D Web, the Metaverse.
1
3000
9000
00:37
What does this all mean for us?
2
12000
2000
00:39
What it means is the Web is going to become an exciting place again.
3
14000
5000
00:44
It's going to become super exciting as we transform
4
19000
3000
00:47
to this highly immersive and interactive world.
5
22000
4000
00:51
With graphics, computing power, low latencies,
6
26000
3000
00:54
these types of applications and possibilities
7
29000
3000
00:57
are going to stream rich data into your lives.
8
32000
5000
01:02
So the Virtual Earth initiative, and other types of these initiatives,
9
37000
5000
01:07
are all about extending our current search metaphor.
10
42000
6000
01:13
When you think about it, we're so constrained by browsing the Web,
11
48000
3000
01:16
remembering URLs, saving favorites.
12
51000
3000
01:19
As we move to search, we rely on the relevance rankings,
13
54000
3000
01:22
the Web matching, the index crawling.
14
57000
3000
01:25
But we want to use our brain!
15
60000
2000
01:27
We want to navigate, explore, discover information.
16
62000
3000
01:30
In order to do that, we have to put you as a user back in the driver's seat.
17
65000
5000
01:35
We need cooperation between you and the computing network and the computer.
18
70000
4000
01:39
So what better way to put you back in the driver's seat
19
74000
4000
01:43
than to put you in the real world that you interact in every day?
20
78000
3000
01:46
Why not leverage the learnings that you've been learning your entire life?
21
81000
4000
01:50
So Virtual Earth is about starting off
22
85000
3000
01:53
creating the first digital representation, comprehensive, of the entire world.
23
88000
5000
01:58
What we want to do is mix in all types of data.
24
93000
3000
02:01
Tag it. Attribute it. Metadata. Get the community to add local depth,
25
96000
5000
02:06
global perspective, local knowledge.
26
101000
3000
02:09
So when you think about this problem,
27
104000
2000
02:11
what an enormous undertaking. Where do you begin?
28
106000
4000
02:15
Well, we collect data from satellites, from airplanes,
29
110000
4000
02:19
from ground vehicles, from people.
30
114000
3000
02:22
This process is an engineering problem,
31
117000
5000
02:27
a mechanical problem, a logistical problem, an operational problem.
32
122000
4000
02:31
Here is an example of our aerial camera.
33
126000
2000
02:33
This is panchromatic. It's actually four color cones.
34
128000
3000
02:36
In addition, it's multi-spectral.
35
131000
2000
02:38
We collect four gigabits per second of data,
36
133000
4000
02:42
if you can imagine that kind of data stream coming down.
37
137000
2000
02:44
That's equivalent to a constellation of 12 satellites at highest res capacity.
38
139000
6000
02:50
We fly these airplanes at 5,000 feet in the air.
39
145000
4000
02:54
You can see the camera on the front. We collect multiple viewpoints,
40
149000
3000
02:57
vantage points, angles, textures. We bring all that data back in.
41
152000
6000
03:03
We sit here -- you know, think about the ground vehicles, the human scale --
42
158000
4000
03:07
what do you see in person? We need to capture that up close
43
162000
2000
03:09
to establish that what it's like-type experience.
44
164000
4000
03:13
I bet many of you have seen the Apple commercials,
45
168000
4000
03:17
kind of poking at the PC for their brilliance and simplicity.
46
172000
6000
03:23
So a little unknown secret is --
47
178000
2000
03:25
did you see the one with the guy, he's got the Web cam?
48
180000
4000
03:29
The poor PC guy. They're duct taping his head. They're just wrapping it on him.
49
184000
4000
03:33
Well, a little unknown secret is his brother actually works on the Virtual Earth team.
50
188000
4000
03:37
(Laughter). So they've got a little bit of a sibling rivalry thing going on here.
51
192000
5000
03:42
But let me tell you -- it doesn't affect his day job.
52
197000
2000
03:44
We think a lot of good can come from this technology.
53
199000
3000
03:47
This was after Katrina. We were the first commercial fleet of airplanes
54
202000
4000
03:51
to be cleared into the disaster impact zone.
55
206000
3000
03:54
We flew the area. We imaged it. We sent in people. We took pictures of interiors,
56
209000
5000
03:59
disaster areas. We helped with the first responders, the search and rescue.
57
214000
4000
04:03
Often the first time anyone saw what happened to their house was on Virtual Earth.
58
218000
5000
04:08
We made it all freely available on the Web, just to --
59
223000
2000
04:10
it was obviously our chance of helping out with the cause.
60
225000
4000
04:14
When we think about how all this comes together,
61
229000
3000
04:17
it's all about software, algorithms and math.
62
232000
4000
04:21
You know, we capture this imagery but to build the 3-D models
63
236000
3000
04:24
we need to do geo-positioning. We need to do geo-registering of the images.
64
239000
5000
04:29
We have to bundle adjust them. Find tie points.
65
244000
2000
04:31
Extract geometry from the images.
66
246000
3000
04:34
This process is a very calculated process.
67
249000
4000
04:38
In fact, it was always done manual.
68
253000
1000
04:39
Hollywood would spend millions of dollars to do a small urban corridor
69
254000
4000
04:43
for a movie because they'd have to do it manually.
70
258000
3000
04:46
They'd drive the streets with lasers called LIDAR.
71
261000
2000
04:48
They'd collected information with photos. They'd manually build each building.
72
263000
4000
04:52
We do this all through software, algorithms and math --
73
267000
2000
04:54
a highly automated pipeline creating these cities.
74
269000
3000
04:57
We took a decimal point off what it cost to build these cities,
75
272000
3000
05:00
and that's how we're going to be able to scale this out and make this reality a dream.
76
275000
4000
05:04
We think about the user interface.
77
279000
2000
05:06
What does it mean to look at it from multiple perspectives?
78
281000
3000
05:09
An ortho-view, a nadir-view. How do you keep the precision of the fidelity of the imagery
79
284000
5000
05:14
while maintaining the fluidity of the model?
80
289000
4000
05:18
I'll wrap up by showing you the --
81
293000
2000
05:20
this is a brand-new peek I haven't really shown into the lab area of Virtual Earth.
82
295000
4000
05:24
What we're doing is -- people like this a lot,
83
299000
3000
05:27
this bird's eye imagery we work with. It's this high resolution data.
84
302000
3000
05:30
But what we've found is they like the fluidity of the 3-D model.
85
305000
4000
05:34
A child can navigate with an Xbox controller or a game controller.
86
309000
4000
05:38
So here what we're trying to do is we bring the picture and project it into the 3-D model space.
87
313000
5000
05:43
You can see all types of resolution. From here, I can slowly pan the image over.
88
318000
6000
05:49
I can get the next image. I can blend and transition.
89
324000
3000
05:52
By doing this I don't lose the original detail. In fact, I might be recording history.
90
327000
5000
05:57
The freshness, the capacity. I can turn this image.
91
332000
3000
06:00
I can look at it from multiple viewpoints and angles.
92
335000
3000
06:03
What we're trying to do is build a virtual world.
93
338000
3000
06:06
We hope that we can make computing a user model you're familiar with,
94
341000
5000
06:11
and really derive insights from you, from all different directions.
95
346000
4000
06:15
I thank you very much for your time.
96
350000
2000
06:17
(Applause)
97
352000
1000
ABOUT THE SPEAKER
Stephen Lawler - General manager of Microsoft's Virtual EarthStephen Lawler and the Virtual Earth team have created an addictively interactive 3D world that is poised to reinvent our view of advertising, gaming, weather/traffic reporting, instant messaging and more.
Why you should listen
Microsoft's Stephen Lawler offers a tour of Virtual Earth that not only reveals the power and potential of the software itself, but also gives a global glimpse of the new virtual frontier of digital globes, the 3D Web and the metaverse.
Lawler also explores the enormous effort it takes to create the fluid blending and shifting between the multiple views and resolutions of Virtual Earth. From the satellites and airplanes that gather photo data for a top-down view to the ground vehicles and headgear-wearing pedestrians who canvas the ground for an eye-level perspective -- all of it represents a monumental effort of logistics and mechanics.
Stephen Lawler | Speaker | TED.com