ABOUT THE SPEAKER
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.

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?

More profile about the speaker
Blaise Agüera y Arcas | Speaker | TED.com
TED2010

Blaise Agüera y Arcas: Augmented-reality maps

Biaise Aguera y Arcas:演示为地图增加现实内容

Filmed:
1,856,682 views

于TED2010的演讲中, Blaise Aguera y Arcas令人屏气地演示了微软的新颖地图技术,为地图增加现实内容的技术。
- 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

Double-click the English transcript below to play the video.

00:15
About a year and a half ago,
0
0
2000
大约于一年半前
00:17
Stephen斯蒂芬 Lawler劳勒, who also gave a talk
1
2000
3000
Stephen Lawler 在TED 2007 做了一个
00:20
here at TEDTED in 2007 on Virtual虚拟 Earth地球,
2
5000
2000
关于虚拟地球的演讲
00:22
brought me over to become成为 the architect建筑师 of Bing Maps地图,
3
7000
4000
也是他使我成为Bing地图的软件建筑师
00:26
which哪一个 is Microsoft's微软的 online-mapping在线地图 effort功夫.
4
11000
3000
那是微软于在线地图方面的一次努力
00:29
In the past过去 two and a half, we've我们已经 been very hard at work
5
14000
2000
过去的两年半中, 我们工作得很努力
00:31
on redefining重新定义 the way maps地图 work online线上.
6
16000
4000
我们试图从新定义地图是如何工作的
00:35
And we really are seeing眼看 this in very different不同 terms条款
7
20000
2000
我们是真正的将它当作一个完全不同的名词来看待
00:37
from the kind of mapping制图 and direction方向 site现场
8
22000
2000
从而改变大家曾经对地图和
00:39
that one is used to.
9
24000
2000
方向的概念
00:41
So, the first thing that you might威力 notice注意 about the mapping制图 site现场
10
26000
2000
所以, 这个地图网页首先引起你注意的
00:43
is just the fluidity流动性 of the zooming缩放 and the panning摇摄,
11
28000
2000
是流畅的地图缩放和平移,
00:45
which哪一个, if you're familiar at all with Seadragon海龙,
12
30000
2000
如果你熟悉Seadragon
00:47
that's where it comes from.
13
32000
2000
我们用的就是它的技术
00:49
Mapping制图 is, of course课程,
14
34000
2000
当然,地图并不仅仅
00:51
not just about cartography制图,
15
36000
2000
涉及制图学
00:53
it's also about imagery意象.
16
38000
2000
它也内含影像
00:55
So, as we zoom-in放大 beyond a certain某些 level水平
17
40000
3000
所以当我们放大到一定的程度
00:58
this resolves做出决议 into a kind of Sim City-like市样
18
43000
3000
这就转换成了虚拟的, 成45度角的
01:01
virtual虚拟 view视图 at 45 degrees.
19
46000
2000
模拟城市
01:03
This can be viewed观看 from any of the cardinal枢机主教 directions方向
20
48000
2000
你可以从任何角度来看它
01:05
to show显示 you the 3D structure结构体 of the city, all the facades外墙.
21
50000
3000
它从各个角度向人们展示了城市的三维机构
01:08
Now, we see this space空间, this three-dimensional三维 environment环境,
22
53000
6000
现在, 我们看到这个空间,这个三维的环境
01:14
as being存在 a canvas帆布 on which哪一个
23
59000
3000
作为一块画布
01:17
all sorts排序 of applications应用 can play out,
24
62000
6000
各种各样的应用程序都可以在上面进行发挥
01:23
and map's地图的 directions方向 are really just one of them.
25
68000
3000
用地图来指引方向只是其中的一个
01:26
If you click点击 on this, you'll你会 see
26
71000
3000
如果你点击这里
01:29
some of the ones那些 that we've我们已经 put out, just in the past过去 couple一对 of months个月
27
74000
3000
你可以看见我们于过去几个月里推出的东西
01:32
since以来 we've我们已经 launched推出.
28
77000
2000
我们的项目已经投放使用
01:34
So, for example, a couple一对 of days after the disaster灾害 in Haiti海地,
29
79000
2000
比如,在海地受灾后的几天里
01:36
we had an earthquake地震 map地图 that showed显示
30
81000
2000
我们的地震地图显示了
01:38
before and after pictures图片 from the sky天空.
31
83000
2000
震前和震后的空拍图片
01:40
This wonderful精彩 one which哪一个 I don't have time to show显示 you
32
85000
2000
由于时间问题,这个精彩的应用我就没办法演示给你们看
01:42
is taking服用 hyper-local超本地 blogs博客 in real真实 time
33
87000
3000
它于第一时间从本地的博客里提取内容
01:45
and mapping制图 those stories故事, those entries
34
90000
2000
将博客中提到的故事和记录
01:47
to the places地方 that are referred简称 to on the blogs博客.
35
92000
2000
用地图来标识
01:49
It's wonderful精彩.
36
94000
3000
真是很精彩
01:52
But I'm going to show显示 you some more candy糖果 sort分类 of stuff东东.
37
97000
3000
然而我要给你们看的是类似眼睛糖果的东西
01:55
So, we see the imagery意象, of course课程,
38
100000
4000
当然,我们看到图象
01:59
not stopping停止 at the sky天空.
39
104000
3000
不仅仅是停留在空中
02:02
These little green绿色 bubbles泡泡 represent代表
40
107000
3000
这些绿色的小泡泡代表
02:05
photosynthsphotosynths that users用户 have made制作.
41
110000
2000
用户做的照相的合成。
02:07
I'm not going to dive潜水 into them either, but photosynthsphotosynths are integrated集成 into the map地图.
42
112000
3000
对此我也不深入展开了,接下来我将展示照片的合成,及融入地图
02:10
Everything that's cased套管 in blue蓝色
43
115000
2000
所有的蓝色地带
02:12
is an area where we've我们已经 taken采取 imagery意象 on the ground地面 as well.
44
117000
5000
都是我们于地上拍过照的地方
02:17
And so, when you fly down --
45
122000
2000
因此, 你往下飞--
02:19
(Applause掌声)
46
124000
2000
(掌声)--
02:21
Thank you. When you fly down to the ground地面,
47
126000
2000
谢谢, 当你飞到陆地的时候
02:23
and you see this kind of panoramic全景 imagery意象,
48
128000
3000
你便可以看见类似的全景图像
02:26
the first thing that you might威力 notice注意 is that it's not just a picture图片,
49
131000
4000
你第一个注意到的是,这只不过是一张照片
02:30
there's just as much three-dimensional三维 understanding理解 of this environment环境
50
135000
3000
但是内含的三维信息
02:33
as there is of the three-dimensional三维 city from above以上,
51
138000
3000
和俯瞰三维城市一样丰富
02:36
so if I click点击 on something to get a closer接近 view视图 of it,
52
141000
4000
所以当我点击目标靠近观察
02:40
then, the fact事实 that that transition过渡 looks容貌 as it does,
53
145000
3000
事实上你可看到的那个转换的过程
02:43
is a function功能 of all of that geometry几何,
54
148000
2000
这都靠几何的功能
02:45
all of that 3D understanding理解 behind背后 this model模型.
55
150000
4000
这个模型的背后是对三维的整体认识
02:49
Now, I'll show显示 you a fun开玩笑 app应用
56
154000
3000
现在,我为你们展示一个有趣的程序
02:52
that -- we've我们已经 been working加工 on a collaboration合作 with our friends朋友 at FlickrFlickr的.
57
157000
7000
是我们伙同Flicker的朋友开发的
02:59
This takes FlickrFlickr的, georegisteredgeoregistered imagery意象
58
164000
3000
这个程序将Flickr里登记过地理位置图形
03:02
and uses使用 photosynth-likePhotosynth的样 processes流程
59
167000
3000
用类似图象合成的运作方式
03:05
to connect that imagery意象 to our imagery意象, so --
60
170000
4000
将那些图片跟我们的图片连接
03:09
I'm not sure if that's the one I actually其实 meant意味着 to pull up, but --
61
174000
3000
我不知道这是我要找出来的东西
03:12
(Laughter笑声)
62
177000
3000
(笑声)
03:15
But notice注意 -- this is, of course课程, a popular流行 tourist游客 site现场,
63
180000
4000
但是,你们可以看到, 这是一个热门的旅游点
03:19
and there are lots of photos相片 around here,
64
184000
2000
这里有很多照片
03:21
and these photos相片 are all taken采取 at different不同 times.
65
186000
3000
这些照片是于不同时候拍的
03:24
So this one was taken采取 around five.
66
189000
2000
这张是早上5点拍的
03:26
So that's the FlickrFlickr的 photo照片,
67
191000
2000
那是Flickr的照片
03:28
that's our imagery意象.
68
193000
2000
这是我们的照片
03:30
So you really see how this kind of crowd-sourced众包 imagery意象
69
195000
3000
这样你可以看见这种群集的图象
03:33
is integrating整合, in a very deep way, into the map地图 itself本身.
70
198000
3000
是如何合成并深层地融入地图中
03:36
(Applause掌声)
71
201000
2000
(掌声)
03:38
Thank you.
72
203000
2000
谢谢
03:40
(Applause掌声)
73
205000
2000
(掌声)
03:42
There are several一些 reasons原因 why this is interesting有趣
74
207000
2000
这之所以很有趣,有几个原因
03:44
and one of them, of course课程, is time travel旅行.
75
209000
2000
其中之一,是时间旅行
03:46
And I'm not going to show显示 you some of the wonderful精彩 historic历史性 imagery意象 in here,
76
211000
2000
我就不展示精彩的历史图象
03:48
but there are some with horses马匹 and carriages车厢 and so on as well.
77
213000
3000
但是其中有马车之类的东西
03:51
But what's cool about this is that, not only is it augmenting增广
78
216000
3000
但是这之所以很酷, 并不仅仅是因为
03:54
this visual视觉 representation表示 of the world世界
79
219000
2000
它充实了对大千世界的视觉表达
03:56
with things that are coming未来 in from users用户,
80
221000
3000
它们从很多人那里汇总再过来
03:59
but it also is the foundation基础 for augmented增强 reality现实,
81
224000
4000
但同时, 这又是增强现实的基础
04:03
and that's something that I'll be showing展示 you more of in just a moment时刻.
82
228000
3000
过一会儿我将更多地向你们介绍这项功能
04:06
Now I just made制作 a transition过渡 indoors在室内. That's also interesting有趣.
83
231000
4000
现在我正在转换进入室内。这也很有趣
04:10
OK, notice注意 there's now a roof屋顶 above以上 us.
84
235000
2000
请注意我们头上有屋顶
04:12
We're inside the Pike梭子鱼 Place地点 Market市场.
85
237000
3000
我们现在是在派克市场里面
04:15
And this is something that we're able能够 to do with a backpack背包 camera相机,
86
240000
2000
这个只要在背包上放一个摄像机就能实现了
04:17
so, we're now not only imaging成像 in the street
87
242000
4000
所以, 我们不光是用汽车顶上
04:21
with this camera相机 on tops上衣 of cars汽车,
88
246000
2000
的摄像机对大街成像
04:23
but we're also imaging成像 inside.
89
248000
4000
而且我们也能对室内成像
04:27
And from here, we're able能够 to do the same相同 sorts排序 of registration注册,
90
252000
5000
从这里, 我们可以使用各种东西成像
04:32
not only of still images图片, but also of video视频.
91
257000
4000
不单是用静态的图象,还可以用录像
04:36
So this is something that we're now going to try
92
261000
2000
我们现在就来试验一下
04:38
for the first time, live生活,
93
263000
2000
这是我们第一次做直播测试
04:40
and this is really, truly, very frightening可怕的.
94
265000
3000
这真是,真的,非常可怕
04:43
(Laughter笑声)
95
268000
2000
(笑声)
04:46
OK.
96
271000
2000
04:48
(Ringing铃声)
97
273000
2000
(电话铃声)
04:50
All right, guys, are you there?
98
275000
2000
好了,伙计们, 你们在吗
04:52
(Noise噪声)
99
277000
2000
(噪音)
04:54
All right. I'm hitting it. I'm punching冲孔 play.
100
279000
2000
好,我上线了,我开始玩了
04:56
I'm live生活. All right. There we go.
101
281000
2000
直播开始了,好的,开始
04:58
So, these are our friends朋友 in Pike梭子鱼 Place地点 Market市场, the lab实验室.
102
283000
5000
这是我们现正位于派克市场的朋友们,我们的实验室
05:03
(Applause掌声)
103
288000
8000
(掌声)
05:12
So they're broadcasting广播 this live生活.
104
297000
2000
他们现在在现场直播
05:14
OK, George乔治, can you pan back over to the corner market市场?
105
299000
4000
好吧,乔治,你能把镜头移到市场的角落吗?
05:18
Because I want to show显示 points of interest利益.
106
303000
4000
因为我想展示一下几个兴趣点
05:22
No, no. The other way.
107
307000
3000
不,不,另外一个方向
05:25
Yeah, yeah, back to the corner, back to the corner.
108
310000
2000
对了,对了,回到角落,回到角落
05:27
I don't want to see you guys yet然而.
109
312000
3000
我还不想看见你们呢
05:31
OK, OK, back to the corner, back to the corner, back to the corner.
110
316000
6000
好吧,好吧,回到角落,回到角落,回到角落
05:37
OK, never mind心神.
111
322000
2000
好吧, 算了
05:40
What I wanted to show显示 you was these points of interest利益
112
325000
2000
我想要给你们展示的几个兴趣点
05:42
over here on top最佳 of the image图片
113
327000
2000
是在顶上的几个图象
05:44
because what that gives you a sense of is the way,
114
329000
2000
因为这个会让你感觉
05:46
if you're actually其实 on the spot,
115
331000
2000
你好象就在当地
05:48
you can think about this --
116
333000
2000
你可以这样想:
05:50
this is taking服用 a step in addition加成 to augmented增强 reality现实.
117
335000
3000
这是在集结现实的基础上又迈进了一步
05:53
What the hell地狱 are you guys -- oh, sorry.
118
338000
2000
你们这些家伙在干什么啊?噢,对不起
05:55
(Laughter笑声)
119
340000
3000
(笑声)
05:58
We're doing two different不同 --
120
343000
3000
我们正在做两件不同的事情
06:01
OK, I'm hanging up now.
121
346000
3000
好吧,我挂了
06:04
We're doing two different不同 things here.
122
349000
2000
我们正在做两件不一样的事情
06:06
One of them is to take that real真实 ...
123
351000
2000
他们中的一个正要去拿那个真的......
06:08
(Laughter笑声)
124
353000
7000
(笑声)
06:15
All right, let me just take a moment时刻 and thank the team球队.
125
360000
3000
好吧, 让我们停下来感谢咱们的小组成员
06:18
They've他们已经 doneDONE a fantastic奇妙 job工作 of pulling this together一起.
126
363000
2000
他们出色地完成了这个实验
06:20
(Applause掌声)
127
365000
2000
(掌声)
06:22
I'm going to abandon放弃 them now and walk步行 back outside.
128
367000
3000
我现在抛弃他们了,接着往外走
06:25
And while I walk步行 outside, I'll just mention提到 that
129
370000
3000
在我往外走的时候,我得提一下的是
06:28
here we're using运用 this for telepresence网真,
130
373000
2000
这里,我们把它用做远程进入现场的工具
06:30
but you can equally一样 well use this
131
375000
2000
但是,你们同时也可以在当地
06:32
on the spot, for augmented增强 reality现实.
132
377000
2000
把这个当做增强现实的工具
06:34
When you use it on the spot, it means手段 that
133
379000
2000
当你在景点现场用的时候
06:36
you're able能够 to bring带来 all of that metadata元数据
134
381000
2000
你就可以把所有的原始数据
06:38
and information信息 about the world世界 to you.
135
383000
2000
和信息都带到你的所在地
06:40
So here, we're taking服用 the extra额外 step of also broadcasting广播 it.
136
385000
2000
而我们进一步做的就是把这些东西转播出去
06:42
That was being存在 broadcast广播, by the way, on a 4G network网络
137
387000
3000
顺便说一下,这是在4G的网络上转播的
06:45
from the market市场.
138
390000
3000
从市场转播的
06:48
All right, and now there's one last TEDTED talk
139
393000
3000
好,现在你们看到的是微软做的最后一个TED讲座
06:51
that Microsoft微软 has given特定 in the past过去 several一些 years年份.
140
396000
2000
这是微软在过去几年里做的一个讲座
06:53
And that's Curtis柯蒂斯 Wong, WorldWide全世界 Telescope望远镜.
141
398000
3000
那是Curtis Wong,万维望远镜
06:56
So, we're going to head over to the dumpsters垃圾箱,
142
401000
2000
我们现在朝垃圾箱走去
06:58
where it's traditional传统, after a long day at the market市场,
143
403000
3000
这也是长时间工作了一天后的传统
07:01
to go out for a break打破, but also stare up at the sky天空.
144
406000
4000
到外面去小歇片刻, 然后注视着天空
07:05
This is the integration积分
145
410000
2000
这是合成
07:07
of WorldWide全世界 Telescope望远镜 into our maps地图.
146
412000
3000
将万维望远镜合成到我们的地图
07:10
(Applause掌声)
147
415000
2000
(掌声)
07:12
This is the current当前 -- thank you --
148
417000
3000
这是现在, 谢谢
07:15
this is the current当前 time. If we scrub擦洗 the time,
149
420000
2000
这是现在,但是如果我们转移时间
07:17
then we can see how the sky天空 will look at different不同 times,
150
422000
3000
那样我们就可以看见天空在不同的时间是如何变化的
07:20
and we can get all of this very detailed详细
151
425000
2000
我们可以得到非常详细的
07:22
information信息 about different不同 times, different不同 dates日期:
152
427000
4000
不同时间,不同日期的信息
07:27
Let's move移动 the moon月亮 a little higher更高 in the sky天空,
153
432000
2000
让我们把月亮在天空中移得高一点
07:29
maybe change更改 the date日期.
154
434000
4000
或者改动日期
07:33
I would like to kind of zoom放大 in on the moon月亮.
155
438000
3000
我想放大月亮的图象
07:39
So, this is an astronomically天文数字 complete完成
156
444000
2000
所以这是将天空完整的
07:41
representation表示 of the sky天空
157
446000
3000
天文代表
07:44
integrated集成 right into the Earth地球.
158
449000
2000
合成到了地球中
07:46
All right now, I've overrun泛滥 my time,
159
451000
2000
好,我的时间到了
07:48
so I've got to stop.
160
453000
2000
我必须要停下了
07:50
Thank you all very much.
161
455000
2000
非常感谢
07:52
(Applause掌声)
162
457000
12000
(鼓掌)
Translated by Jenny Yang
Reviewed by Chaoran Yu

▲Back to top

ABOUT THE SPEAKER
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.

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?

More profile about the speaker
Blaise Agüera y Arcas | Speaker | TED.com