Ed Ulbrich: How Benjamin Button got his face

翻译人员: Yang Wang 校对人员: Yiqing Xu

I'm here today representing a team of artists and technologists and filmmakers

今天我在这里代表了一队艺术家、技术专家和电影制作者

that worked together on a remarkable film project for the last four years.

在过去的四年里，他们在一起为一个了不起的电影项目工作。

And along the way they created a breakthrough in computer visualization.

在这一过程中，他们在计算机视觉特效做出了突破。

So I want to show you a clip of the film now.

所以我现在想为你们播放一段电影片段。

Hopefully it won't stutter.

但愿播放不会卡。

And if we did our jobs well, you won't know that we were even involved.

若是我们做得好的话，你们不会觉察到我们曾参与其中。

Voice (Video): I don't know how it's possible ...

电影内容：我不知道这怎么可能。。。

but you seem to have more hair.

但是看上去你的头发多了。

Brad Pitt: What if I told you that I wasn't getting older ...

本杰明·巴顿：如果我告诉你我没有变老。。。

but I was getting younger than everybody else?

却变得比比人年轻了会怎样？

I was born with some form of disease.

我生来就得了一种病。

Voice: What kind of disease?

声音：什么病？

BP: I was born old.

本杰明·巴顿：我生来年老。

Man: I'm sorry.

男声：抱歉。

BP: No need to be. There's nothing wrong with old age.

本杰明·巴顿：没必要。年老没有什么错。

Girl: Are you sick?

女声：你病了吗？

BP: I heard momma and Tizzy whisper,

本杰明·巴顿：我听到妈妈和Tizzy偷偷说话，

and they said I was gonna die soon.

他们说我就快要死了。

But ... maybe not.

但是。。。或许不会。

Girl: You're different than anybody I've ever met.

女声：你同我见过的所有人都不一样。

BB: There were many changes ...

本杰明·巴顿：变化了好多。。。

some you could see, some you couldn't.

一些你能看到，一些你不能。

Hair started growing in all sorts of places,

毛发开始在身体的各个地方生长，

along with other things.

一些别的东西也是。

I felt pretty good, considering.

我感觉相当好。

Ed Ulbrich: That was a clip from "The Curious Case of Benjamin Button."

Ed Ulbrich：刚刚播放的是来自《本杰明巴顿奇事》的电影片段。

Many of you, maybe you've seen it or you've heard of the story,

你们中的许多人可能已经看过或者听说过这个故事。

but what you might not know

但是你或许有所不知的是

is that for nearly the first hour of the film,

在电影的接近一小时里，

the main character, Benjamin Button, who's played by Brad Pitt,

由布拉德皮特扮演的主角，本杰明巴顿

is completely computer-generated from the neck up.

他脖子以上的部分完全是由计算机生成的。

Now, there's no use of prosthetic makeup

注意，我们没有使用假体化妆

or photography of Brad superimposed over another actor's body.

或者将布拉德的照片附加到另一个演员的身体上。

We've created a completely digital human head.

我们创造了一个完全数字的人类头部。

So I'd like to start with a little bit of history on the project.

现在我想开始讲一点这个项目的历史。

This is based on an F. Scott Fitzgerald short story.

这是基于菲兹杰拉德的一部短篇故事。

It's about a man who's born old and lives his life in reverse.

讲的是一个生下来就很老然后过着相反的生活。

Now, this movie has floated around Hollywood

现在，这个电影已经在好莱坞发行了

for well over half a century,

半个世纪了

and we first got involved with the project in the early '90s,

在九十年代早期，我们第一次参与了这个项目

with Ron Howard as the director.

当时的导演是朗·霍华德

We took a lot of meetings and we seriously considered it.

我们开了很多会并认真的考虑了

But at the time we had to throw in the towel.

但是当时我们不得不投降认输

It was deemed impossible.

当时被认为是不可能的。

It was beyond the technology of the day to depict a man aging backwards.

刻画一个人从老到年轻，这超越了当时的技术能力。

The human form, in particular the human head,

人类的构型，尤其是头部，

has been considered the Holy Grail of our industry.

被看作是我们工业里的圣杯。

The project came back to us about a decade later,

十年后这个项目由回到我们这里，

and this time with a director named David Fincher.

这一回导演的名字是大卫·芬奇

Now, Fincher is an interesting guy.

芬奇是个有趣的家伙。

David is fearless of technology,

他无惧于技术，

and he is absolutely tenacious.

并且绝对固执。

And David won't take "no."

并且大卫不会接受“不”。

And David believed, like we do in the visual effects industry,

大卫相信，就像我们对视觉特效工业的观点一样，

that anything is possible

没有不可能

as long as you have enough time, resources and, of course, money.

前提是你有足够的时间，资源，当然，还有钱

And so David had an interesting take on the film,

因此大卫有兴趣接手这部电影，

and he threw a challenge at us.

并且给我们抛来一个挑战。

He wanted the main character of the film to be played

他希望电影中的主角

from the cradle to the grave by one actor.

从摇篮里出来一直到进入坟墓，都由一个演员来演。

It happened to be this guy.

刚好是这个人。

We went through a process of elimination and a process of discovery

我们经历了削减和发现的过程

with David, and we ruled out, of course, swapping actors.

同大卫一道，当然，我们放弃了换演员的想法。

That was one idea: that we would have different actors,

那个想法就是：我们会有不同的演员，

and we would hand off from actor to actor.

并且我们会用一个又一个演员。

We even ruled out the idea of using makeup.

我们划掉了使用化妆的想法。

We realized that prosthetic makeup just wouldn't hold up,

我们意识到假体化妆就是没法做得真。

particularly in close-up.

特别是近看的时候。

And makeup is an additive process. You have to build the face up.

而且化妆是一个额外的步骤。 你不得不在脸的外边来化。

And David wanted to carve deeply into Brad's face

大卫想让皱纹深深的刻在布拉德的脸上

to bring the aging to this character.

以显示出这个角色的年纪变化。

He needed to be a very sympathetic character.

他需要一个非常的意气相投的人物。

So we decided to cast a series of little people

所以我们决定拍摄一系列的小人物

that would play the different bodies of Benjamin

他们扮演本杰明

at the different increments of his life

在不同时候的不同的身体

and that we would in fact create a computer-generated version of Brad's head,

并且我们事实上会创造一个计算机生成的布拉德的头部。

aged to appear as Benjamin,

随本杰明年纪而变化，

and attach that to the body of the real actor.

并将它附到真的演员身体上。

Sounded great.

听上去很棒。

Of course, this was the Holy Grail of our industry,

当然，这曾经是我们这一行离得圣杯，

and the fact that this guy is a global icon didn't help either,

事实上布拉德作为一个全球符号也没能帮上一把，

because I'm sure if any of you ever stand in line at the grocery store,

因为我很确信如果你们任何人站成一行在杂货店里，

you know -- we see his face constantly.

你知道，我们总是看到他的脸。

So there really was no tolerable margin of error.

所以这里边没有容错的空间。

There were two studios involved: Warner Brothers and Paramount.

华纳兄弟和派拉蒙参与在其中。

And they both believed this would make an amazing film, of course,

他们都相信这无疑会是一个令人惊奇的电影

but it was a very high-risk proposition.

但是它也是一个非常高风险的提议。

There was lots of money and reputations at stake.

很多的钱与声誉都处在危险关头。

But we believed that we had a very solid methodology

我们相信我们有一个非常坚实的方法

that might work ...

可能管用。。。

But despite our verbal assurances,

但是无论我们如何口头承诺，

they wanted some proof.

他们要一些证明。

And so, in 2004, they commissioned us to do a screen test of Benjamin.

为此，在2004年，他们委托我们做一次本杰明的试镜。

And we did it in about five weeks.

我们差不多五周后做给他们看。

But we used lots of cheats and shortcuts.

但是我们在其中参了好多假并走了捷径。

We basically put something together to get through the meeting.

基本上我们把一些东西搁了一块儿就通过了会议。

I'll roll that for you now. This was the first test for Benjamin Button.

我现在把那些拿出来给你们看。 这是本杰明巴顿的第一个测试。

And in here, you can see, that's a computer-generated head --

然后在这里，你肯一看到，那是一个计算机生成的头部。

pretty good -- attached to the body of an actor.

相当不错。附到演员的身体后的样子。

And it worked. And it gave the studio great relief.

这个成了。 制片厂松了一口气。

After many years of starts and stops on this project,

在这个项目拍拍停停好多年，

and making that tough decision,

并做了艰难的选择后，

they finally decided to greenlight the movie.

他们最终决定为这个电影开绿灯。

And I can remember, actually, when I got the phone call to congratulate us,

我可以记起来，事实上，但我收到祝贺我们的电话，

to say the movie was a go,

说电影要拍了的时候，

I actually threw up.

我其实是一团糟。

(Laughter)

（笑声）

You know, this is some tough stuff.

你知道，这是一些艰难的东西。

So we started to have early team meetings,

然后我们开始了早期的团队会议，

and we got everybody together,

并且把所有人聚起来，

and it was really more like therapy in the beginning,

开始的时候更像是治疗，

convincing each other and reassuring each other that we could actually undertake this.

得相互说服、保证对方说我们确实能承担这些。

We had to hold up an hour of a movie with a character.

我们得支撑一部单人物持续一小时的电影

And it's not a special effects film; it has to be a man.

而且他并不是一部特效电影；中间得有一个人。

We really felt like we were in a -- kind of a 12-step program.

我们觉得自己就好像是在---某种戒瘾过程中。

And of course, the first step is: admit you've got a problem. (Laughter)

当然，第一步是：承认自己有一个问题。

So we had a big problem:

我们有一个大问题。

we didn't know how we were going to do this.

我们不知道我们该怎样去做。

But we did know one thing.

但我们知道一件事。

Being from the visual effects industry,

在视觉特效工业里，

we, with David, believed that we now had enough time,

我们，连同大卫，相信我们现在有足够的时间，

enough resources, and, God, we hoped we had enough money.

足够的资源，然后，天哪，我们希望我们有足够的钱。

And we had enough passion to will the processes and technology into existence.

并且我们有足够的激情去用意志趋势工序与技术变成实际。

So, when you're faced with something like that,

所以，当你面对一些这样的事情时，

of course you've got to break it down.

当然你会把它分块儿。

You take the big problem and you break it down into smaller pieces

你取一个大问题然后把它分成几个小问题

and you start to attack that.

然后开始攻破。

So we had three main areas that we had to focus on.

我们有三个主要的需要专注的地方。

We needed to make Brad look a lot older --

我们需要让布拉德看上去老许多。

needed to age him 45 years or so.

我们需要他差不多45岁的样子。

And we also needed to make sure that we could take Brad's idiosyncrasies,

我们也需要确保我们可以抓住布拉德的特殊癖好行为，

his little tics, the little subtleties that make him who he is

他的小抽搐，那些能体现他的细微之处

and have that translate through our process

并将它们转移到我们的处理过程中

so that it appears in Benjamin on the screen.

使得它出现在银幕上的本杰明中。

And we also needed to create a character

我们还需要创造一个角色

that could hold up under, really, all conditions.

他可以支撑起所有情况。

He needed to be able to walk in broad daylight,

他得能够行走在大白天，

at nighttime, under candlelight,

晚上，烛光中，

he had to hold an extreme close-up,

他得能够撑住大特写镜头，

he had to deliver dialogue,

他得能够传递对话，

he had to be able to run, he had to be able to sweat,

他得会跑，出汗，

he had to be able to take a bath, to cry,

他还得会洗澡，哭泣，

he even had to throw up.

他甚至得会呕吐。

Not all at the same time --

当然不是同时做这些事情。

but he had to, you know, do all of those things.

但是他得会，你知道，做所有这些事情。

And the work had to hold up for almost the first hour of the movie.

并且这些工作得支撑电影的差不多前一个小时。

We did about 325 shots.

我们拍摄了325个镜头

So we needed a system that would allow Benjamin

因此我们需要一个系统能够让本杰明

to do everything a human being can do.

去做任何人可以做的事情。

And we realized that there was a giant chasm

我们意识到一个大分歧

between the state of the art of technology in 2004

出现在2004年的技术艺术水平

and where we needed it to be.

和我们的愿望之间。

So we focused on motion capture.

因此我们集中在运动捕捉。

I'm sure many of you have seen motion capture.

我相信你们中许多人已经看过运动捕捉了。

The state of the art at the time

当时的工艺水平

was something called marker-based motion capture.

叫做基于标记的动作捕捉。

I'll give you an example here.

这里我给你们举一个例子。

It's basically the idea of, you wear a leotard,

想法基本上就是，你穿一个连衣裤，

and they put some reflective markers on your body,

然后安放一些反光标志

and instead of using cameras,

用来替换摄像机的，

there're infrared sensors around a volume,

是围绕在一个空间内的红外感应传感器

and those infrared sensors track the three-dimensional position

那些红外传感器实时地跟踪

of those markers in real time.

那些标记的空间位置

And then animators can take the data of the motion of those markers

然后动画师收集这些数据

and apply them to a computer-generated character.

并应用到计算机生成的人物中。

You can see the computer characters on the right

你可以看到右边的计算机人物

are having the same complex motion as the dancers.

正做着着同舞蹈者相同的复杂运动。

But we also looked at numbers of other films at the time

与此同时我们也关注其他的

that were using facial marker tracking,

运用脸部捕捉技术的电影

and that's the idea of putting markers on the human face

将标志放在脸部

and doing the same process.

进行同样的处理

And as you can see, it gives you a pretty crappy performance.

正如你所看到的，这种技术所生成的画面相当糟糕

That's not terribly compelling.

这并不十分引人注目。

And what we realized

我们意识到

was that what we needed

我们所需要的

was the information that was going on between the markers.

是那些在这些标志点之间的信息

We needed the subtleties of the skin.

我们需要皮肤的细节

We needed to see skin moving over muscle moving over bone.

我们需要透过皮肤看到肌肉和骨骼的运动

We needed creases and dimples and wrinkles and all of those things.

我们需要皱纹，酒窝之类的

Our first revelation was to completely abort and walk away from

我们第一件要做的事就是抛弃现今所用的技术

the technology of the day, the status quo, the state of the art.

这一天的技术，现状，最先进的。

So we aborted using motion capture.

所以我们放弃了动作捕捉

And we were now well out of our comfort zone,

我们离开了我们所熟悉的领域

and in uncharted territory.

进入了从未涉及过的新领域

So we were left with this idea

我们进入了一个技术上的难点

that we ended up calling "technology stew."

我们结束了通话“技术炖。”

We started to look out in other fields.

我们开始在其他领域需求帮助

The idea was that we were going to find

我们相信我们会找到我们所需要的技术

nuggets or gems of technology

金块或技术宝石

that come from other industries like medical imaging,

这项技术可能是从其他领域， 例如医药

the video game space,

或是家庭游戏机

and re-appropriate them.

或是重新调整他们

And we had to create kind of a sauce.

或者我们需要一项全新的技术

And the sauce was code in software

这项技术可以编写成一个软件

that we'd written to allow these disparate pieces of technology

允许我们将这些零散的信息组合在一起

to come together and work as one.

走到一起，齐心协力。

Initially, we came across some remarkable research

起初，我们遇见一些研究

done by a gentleman named Dr. Paul Ekman in the early '70s.

由Paul在70年代早期得到的

He believed that he could, in fact,

他认为 他可以将人类的面部表情归类

catalog the human face.

目录中的人脸。

And he came up with this idea of Facial Action Coding System, or FACS.

他设计了面部动作编码系统

He believed that there were 70 basic poses

他认为有10个基本的面部表情

or shapes of the human face,

人类面对或形状

and that those basic poses or shapes of the face

借由这些动作去构成无数种可能

can be combined to create infinite possibilities

它们可以结合起来，创造无限的可能性

of everything the human face is capable of doing.

一切人类面对的就是这样做的能力。

And of course, these transcend age, race, culture, gender.

这些全部不受年龄 种族 文化 性别的限制

So this became the foundation of our research as we went forward.

这成为我们继续的基石

And then we came across some remarkable technology

即接着我们遇到了一些卓越的科技

called Contour.

被称作Contour

And here you can see a subject having phosphorus makeup

你可以看到这个人的面部 具有磷 标记

stippled on her face.

点刻在她的脸上。

And now what we're looking at is really creating a surface capture

我们现在说看到的是一个真实面部捕捉

as opposed to a marker capture.

而不是那种标记捕捉

The subject stands in front of a computer array of cameras,

它站于一组由电脑驱动的相机阵列前

and those cameras can, frame-by-frame,

这些相机将会准确的描绘出物体的运动

reconstruct the geometry of exactly what the subject's doing at the moment.

重建几何正是这一主题的目前做的事情。

So, effectively, you get 3D data in real time of the subject.

我们可一有效的获得物体在3D空间的运动数据

And if you look in a comparison,

对比下这两种技术

on the left, we see what volumetric data gives us

在左侧，我们看到的数据给了我们什么容积

and on the right you see what markers give us.

并在右边你看到什么标记给我们。

So, clearly, we were in a substantially better place for this.

显而易见，这种技术是多么优越了

But these were the early days of this technology,

但是这还只是这项技术的雏形，

and it wasn't really proven yet.

还没有被完全证实。

We measure complexity and fidelity of data

但是，我们可以对比多边形定点个数做个比较。

in terms of polygonal count.

在多边形数计算。

And so, on the left, we were seeing 100,000 polygons.

左侧的是100，000个顶点

We could go up into the millions of polygons.

右边这个却已经打到了上亿个顶点

It seemed to be infinite.

或者可以认为是无限个

This was when we had our "Aha!"

这令我们豁然开朗

This was the breakthrough.

这是我们这个项目的一个中重大突破啊

This is when we're like, "OK, we're going to be OK,

这令我们有一种我们可以完成这部电影的信心了

This is actually going to work."

这一定会成功的

And the "Aha!" was, what if we could take Brad Pitt,

和“一哈！”是，如果我们可以采取布拉德皮特，

and we could put Brad in this device,

我们可以把布拉德本装置，

and use this Contour process,

并利用这一轮廓的过程中，

and we could stipple on this phosphorescent makeup

我们可以在这个点画磷光化妆

and put him under the black lights,

并把他下黑灯，

and we could, in fact, scan him in real time

我们可以，事实上，他在实时扫描

performing Ekman's FACS poses.

表演埃克曼的流式细胞仪构成。

Right? So, effectively,

看 非常有效

we ended up with a 3D database

我们最终完成了一个3D数据库

of everything Brad Pitt's face is capable of doing.

包含了所有Brad Pitt 的所有表情

(Laughter)

大笑

From there, we actually carved up those faces

这里， 我们实际上是分割了这些面部表情

into smaller pieces and components of his face.

分解为许多组成部分

So we ended up with literally thousands and thousands and thousands of shapes,

我们获得了上千种形状

a complete database of all possibilities

一个他面部所能做到的各种表情的数据库

that his face is capable of doing.

他面对的就是这样做的能力。

Now, that's great, except we had him at age 44.

这一切看起来都很不错， 不过这是他44岁的模样

We need to put another 40 years on him at this point.

我们需要将这个年龄加上40岁

We brought in Rick Baker,

我们于是请来了Rick Baker,

and Rick is one of the great makeup and special effects gurus

他是一名强力的化妆和特效天才，在电影行业中。

of our industry.

我们的工业。

And we also brought in a gentleman named Kazu Tsuji,

我们还有Kazu suji

and Kazu Tsuji is one of the great photorealist sculptors of our time.

Kazu 是一名非常好的动态光影建模师

And we commissioned them to make a maquette,

我们的人任务就是做一个模型

or a bust, of Benjamin.

或是 毁了他 本杰明

So, in the spirit of "The Great Unveiling" -- I had to do this --

因此，在“大揭幕精神” - 我不得不这样做

I had to unveil something.

我已经推出的东西。

So this is Ben 80.

这是ben 80岁的样子

We created three of these:

我们做了三个模型

there's Ben 80, there's Ben 70, there's Ben 60.

分别是本 80 70 60

And this really became the template for moving forward.

这是我们继续的模板

Now, this was made from a life cast of Brad.

现在，这是从一个生命铸造的布拉德。

So, in fact, anatomically, it is correct.

因此，事实上，在解剖学上，这是正确的。

The eyes, the jaw, the teeth:

的眼睛，下巴，牙齿...

everything is in perfect alignment with what the real guy has.

一切都是在完美的路线是什么真正的家伙。

We have these maquettes scanned into the computer

我们这些maquettes扫描进电脑

at very high resolution --

在非常高的分辨率。

enormous polygonal count.

巨大的多边形数。

And so now we had three age increments of Benjamin

所以现在我们有三个年龄递增本杰明

in the computer.

在计算机中。

But we needed to get a database of him doing more than that.

但是，我们必须找到他的数据库做不止于此。

We went through this process, then, called retargeting.

我们经历了这一过程，那么，所谓的重新定向。

This is Brad doing one of the Ekman FACS poses.

这是布拉德埃克曼流式细胞仪做的一个姿势。

And here's the resulting data that comes from that,

这里的结果数据来自该来了，

the model that comes from that.

该模型从该来。

Retargeting is the process of transposing that data

和重定目标的过程是数据的移调

onto another model.

到另一个模型。

And because the life cast, or the bust -- the maquette -- of Benjamin

而且因为生活投，或胸部 - 在maquette - 本雅明

was made from Brad,

是从布拉德，

we could transpose the data of Brad at 44

我们可以在44转布拉德数据

onto Brad at 87.

在87到布拉德。

So now, we had a 3D database of everything Brad Pitt's face can do

所以，现在，我们有一个三维数据库的一切布拉德皮特的脸可以做

at age 87, in his 70s and in his 60s.

87岁的他70多岁，在他60岁。

Next we had to go into the shooting process.

下一步，我们已进入了拍摄过程。

So while all that's going on,

因此，尽管所有这回事，

we're down in New Orleans and locations around the world.

我们下来在新奥尔良和世界各地的位置。

And we shot our body actors,

我们的身体，我们拍摄的演员，

and we shot them wearing blue hoods.

我们向他们拍摄身穿蓝色头巾。

So these are the gentleman who played Benjamin.

因此，这是绅士玩本杰明。

And the blue hoods helped us with two things:

和蓝盖帮我们两件事情：

one, we could easily erase their heads;

1，我们可以很容易抹掉他们的头上;

and we also put tracking markers on their heads

我们也把他们的头标记跟踪

so we could recreate the camera motion

这样我们就可以重新相机的动作

and the lens optics from the set.

从镜头的光学设置。

But now we needed to get Brad's performance to drive our virtual Benjamin.

但现在我们需要得到布拉德的表现，以推动我们的虚拟本杰明。

And so we edited the footage that was shot on location

因此，我们编辑了这是拍摄画面上的位置

with the rest of the cast and the body actors

与其他演员和演员的身体

and about six months later

约6个月后

we brought Brad onto a sound stage in Los Angeles

我们带来了布拉德到一个在洛杉矶的声场

and he watched on the screen.

他看着屏幕上

His job, then, was to become Benjamin.

当时他的工作成为本杰明。

And so we looped the scenes.

因此，我们循环的场景。

He watched again and again.

他一遍又一遍。

We encouraged him to improvise.

我们鼓励他即兴。

And he took Benjamin into interesting and unusual places

他带成有趣和不寻常的地方本杰明

that we didn't think he was going to go.

我们不认为他会去。

We shot him with four HD cameras

我们打到了他与4高清摄像机

so we'd get multiple views of him

所以我们会得到他的多个视图

and then David would choose the take of Brad being Benjamin

然后戴维会选择利用被本杰明布拉德

that he thought best matched the footage

他认为最匹配的镜头

with the rest of the cast.

同其他演员。

From there we went into a process called image analysis.

从那里我们去到一个名为图像分析过程。

And so here, you can see again, the chosen take.

所以在这里，你可以再次看到，所选择的需要。

And you are seeing, now, that data being transposed on to Ben 87.

我们看到，现在，该数据被调换到本87。

And so, what's interesting about this is

因此，有趣的是这

we used something called image analysis,

我们使用一种叫做图像分析，

which is taking timings from different components of Benjamin's face.

它正在从本雅明的脸不同部分的时间安排。

And so we could choose, say, his left eyebrow.

所以我们可以选择，例如，他的左眼眉。

And the software would tell us that, well,

而且，软件会告诉我们，好，

in frame 14 the left eyebrow begins to move from here to here,

在14帧的左眼眉开始从这里到这里，

and it concludes moving in frame 32.

并得出结论朝框架32。

And so we could choose numbers of positions on the face

因此，我们可以选择的职位数量在脸上

to pull that data from.

拉从数据。

And then, the sauce I talked about with our technology stew --

然后，我谈酱油与我们的技术炖，

that secret sauce was, effectively, software that allowed us to

这是秘密酱油，有效，使我们的软件

match the performance footage of Brad

布拉德性能相匹配的镜头

in live action with our database of aged Benjamin,

与我们的老年生活本杰明数据库的行动，

the FACS shapes that we had.

流式细胞仪的形状，我们有。

On a frame-by-frame basis,

上一帧一帧的基础上，

we could actually reconstruct a 3D head

其实，我们可以重建一个三维头部

that exactly matched the performance of Brad.

这正好符合了布拉德的表现。

So this was how the finished shot appeared in the film.

因此，这是如何拍摄完成的影片中出现。

And here you can see the body actor.

在这里你可以看到演员的身体。

And then this is what we called the "dead head," no reference to Jerry Garcia.

然后，这就是我们所谓的“死头”，不向杰里加西亚参考。

And then here's the reconstructed performance

然后这里的改造企业绩效

now with the timings of the performance.

现在的表现计时。

And then, again, the final shot.

然后，再次，最后一幕。

It was a long process.

这是一个漫长的过程。

(Applause)

（鼓掌）

The next section here, I'm going to just blast through this,

下一节在这里，我要通过这个刚刚爆炸，

because we could do a whole TEDTalk on the next several slides.

因为我们可以做多张幻灯片的下一个整体TEDTalk。

We had to create a lighting system.

我们必须创造一个照明系统。

So really, a big part of our processes was creating a lighting environment

因此，其实我们的很大一部分是在创造过程中的照明环境

for every single location that Benjamin had to appear

对于每一个位置，本杰明已经出现

so that we could put Ben's head into any scene

这样我们可以把任何场景本的头

and it would exactly match the lighting that's on the other actors

它会完全匹配的照明，关于其他方面的

in the real world.

在现实世界中。

We also had to create an eye system.

我们还必须建立一个眼系统。

We found the old adage, you know,

我们发现旧的格言，你知道，

"The eyes are the window to the soul,"

“眼睛是灵魂之，窗口”

absolutely true.

千真万确。

So the key here was to keep everybody looking in Ben's eyes.

因此，这里的关键是保持大家在Ben的眼睛。

And if you could feel the warmth, and feel the humanity,

如果你能感受到温暖，感受到人类，

and feel his intent coming through the eyes,

他的意图和感受来通过眼睛，

then we would succeed.

然后我们会成功。

So we had one person focused on the eye system

因此，我们有一个人对眼睛的聚焦系统

for almost two full years.

几乎整整两年。

We also had to create a mouth system.

我们还必须创建一个口系统。

We worked from dental molds of Brad.

我们还必须创建一个口系统。

We had to age the teeth over time.

我们必须随着时间的推移年龄的牙齿。

We also had to create an articulating tongue that allowed him to enunciate his words.

我们还必须创造一个阐明舌头，让他阐述他的话。

There was a whole system written in software to articulate the tongue.

整个系统有一个软件书面阐明了舌头。

We had one person devoted to the tongue for about nine months.

我们有一个人专门为约9个月的舌头。

He was very popular.

他很受欢迎。

Skin displacement: another big deal.

皮肤位移：另一个大问题。

The skin had to be absolutely accurate.

皮肤必须绝对准确，

He's also in an old age home, he's in a nursing home

而且他也有一老年之家，他在一家养老院

around other old people,

周围的其他老人，

so he had to look exactly the same as the others.

所以他看的人完全一样。

So, lots of work on skin deformation,

因此，大量的工作，对皮肤变形，

you can see in some of these cases it works,

你可以看到在这些情况下，一些作品，

in some cases it looks bad.

在某些情况下，它看起来不错，

This is a very, very, very early test in our process.

这是一个非常，非常，很早就在我们的测试过程。

So, effectively we created a digital puppet

所以，我们创建了一个有效的数字傀儡

that Brad Pitt could operate with his own face.

可经营的布拉德皮特与他自己的脸。

There were no animators necessary to come in and interpret behavior

有没有必要来在动画和解释行为

or enhance his performance.

或加强他的表现。

There was something that we encountered, though,

有些东西，我们遇到的，虽然，

that we ended up calling "the digital Botox effect."

这要求我们结束了“数字肉毒杆菌的效果。”

So, as things went through this process,

因此，在这个过程中经历的事情，

Fincher would always say, "It sandblasts the edges off of the performance."

芬奇总是说：“这sandblasts边关闭的表现。”

And thing our process and the technology couldn't do,

其中一件我们的工艺和技术不能做

is they couldn't understand intent,

它是无法理解的意图，

the intent of the actor.

该演员意图。

So it sees a smile as a smile.

因此，它认为，作为一个微笑的微笑。

It doesn't recognize an ironic smile, or a happy smile,

它不承认一个具有讽刺意味的微笑，或者一个幸福的微笑，

or a frustrated smile.

或沮丧的笑容。

So it did take humans to kind of push it one way or another.

因此，它确实需要人类推到一个或另一种方式。

But we ended up calling the entire process

但我们最终调用的全过程

and all the technology "emotion capture,"

和所有的技术“情感捕捉”

as opposed to just motion capture.

而不仅仅是动作捕捉。

Take another look.

再来看看。

Brad Pitt: Well, I heard momma and Tizzy whisper,

布拉德皮特：嗯，我听到妈妈和如痴如醉的耳语，

and they said I was gonna die soon,

他们说我是将要死去不久，

but ... maybe not.

但是...也许不会。

EU: That's how to create a digital human in 18 minutes.

埃德尔布里奇：这是如何创建一个数字人体在18分钟。

(Applause)

（掌声）

A couple of quick factoids;

一个快速改编过的夫妇。

it really took 155 people over two years,

他花费了155人超过两年的时间，

and we didn't even talk about 60 hairstyles and an all-digital haircut.

我们甚至没有谈到60发型和全数字理发。

But, that is Benjamin. Thank you.

但是，这是本杰明。谢谢。