Ed Ulbrich: How Benjamin Button got his face

譯者: Adrienne Lin 審譯者: Peiru Liao

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

今天我代表一群藝術家、技術人員、電影製作團隊

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

過去四年間，我們共同完成了一項非凡的電影工程

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

過程中也突破了電影繪圖技術

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

我先讓大家看個電影片段

Hopefully it won't stutter.

希望播放時不會卡住

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

如果我們做好工作的話，你們不會看出我們動了什麼手腳

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

影片: 我不知道這是怎麼回事

but you seem to have more hair.

但你頭髮好像變多了

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

班傑明巴頓: 如果我告訴你，我沒有越來越老

but I was getting younger than everybody else?

反而是越來越年輕了呢?

I was born with some form of disease.

我出生就有一種病

Voice: What kind of disease?

什麼樣的病?

BP: I was born old.

我出生就是老人的樣子

Man: I'm sorry.

我很遺憾

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

不用遺憾，老又沒有錯

Girl: Are you sick?

你生病了嗎?

BP: I heard momma and Tizzy whisper,

我聽到媽和Tizzy講悄悄話

and they said I was gonna die soon.

說我活不了多久

But ... maybe not.

但... 也許他們說錯了

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

我沒見過像你這樣的人

BB: There were many changes ...

我身體開始變化

some you could see, some you couldn't.

有些看的見，有些則不行

Hair started growing in all sorts of places,

體毛開始到處長

along with other things.

還有一些其他東西

I felt pretty good, considering.

其實這感覺還不賴

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

Ed Ulbrich: 這是電影"班傑明的奇幻旅程"片段

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

相信很多人已經看過、或聽過這故事

but what you might not know

但你也許不知道

is that for nearly the first hour of the film,

電影前將近一小時

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

由布萊德彼特飾演的主角 班傑明巴頓

is completely computer-generated from the neck up.

頸部以上完全是電腦合成的

Now, there's no use of prosthetic makeup

沒有靠任何化妝技術

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

也不是布萊德的臉，蓋過另一個演員的身體

We've created a completely digital human head.

我們全以數位方式製造出人頭

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

我先稍微說一下製作起源

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

電影改編自史考特．費滋傑羅的短篇故事

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

關於一個出生時80歲的男人，越活越年輕的故事

Now, this movie has floated around Hollywood

這部電影在好萊塢就謠傳了很久

for well over half a century,

超過50年有吧

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

我們第一次接觸這計劃是在1990年初

with Ron Howard as the director.

導演是朗霍華

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

我們討論了很久，認真考慮要接

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

到後來真的必須放棄的時候

It was deemed impossible.

這已經被認為是不可能的任務

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

當時的科技無法刻畫出返老還童的樣子

The human form, in particular the human head,

整個人，特別是頭的部份

has been considered the Holy Grail of our industry.

在我們這一行，稱作是聖杯傳奇

The project came back to us about a decade later,

約十年後又有人提起這計畫

and this time with a director named David Fincher.

這次的導演是大衛芬奇

Now, Fincher is an interesting guy.

芬奇真的是怪咖一個

David is fearless of technology,

一點都不怕技術上困難

and he is absolutely tenacious.

就是堅持一定要做

And David won't take "no."

沒有人能拒絕他

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

大衛、和我們做特效的持同樣看法

that anything is possible

沒有什麼是不可能的

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

只要有足夠的時間、資源，當然還有資金

And so David had an interesting take on the film,

大衛對這部電影有獨特的想法

and he threw a challenge at us.

丟給我們這項考驗

He wanted the main character of the film to be played

他要這部電影的主角，從出生到死亡

from the cradle to the grave by one actor.

都要是同一個人來演

It happened to be this guy.

也就是這個人

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

我們經過了一連串淘汰、一連串新發現

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

由於大衛堅持，我們排除使用不同演員的方法

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

本來是想利用不同演員

and we would hand off from actor to actor.

分別演出不同階段的人生

We even ruled out the idea of using makeup.

我們也排除利用化妝

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

因為用石膏來弄根本做不出來

particularly in close-up.

尤其是近拍的時候

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

況且也要先做出一張臉才能化

And David wanted to carve deeply into Brad's face

大衛想要在布萊德彼特的臉上刻畫出皺紋

to bring the aging to this character.

來表現出這角色的年老樣貌

He needed to be a very sympathetic character.

他必須是個能讓人感同身受的角色

So we decided to cast a series of little people

所以我們決定選出一批小人物

that would play the different bodies of Benjamin

來演出班傑明在不同人生階段

at the different increments of his life

的各種身體樣貌

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

然後我們利用電腦合成出布萊德的頭

aged to appear as Benjamin,

弄出不同年紀的班傑明

and attach that to the body of the real actor.

在接上其他演員的身體

Sounded great.

聽起來很可行

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

當然，這可是我們這行的聖杯傳奇

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

加上這傢伙沒有人不認識

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

只要去過雜貨店一定看過他的臉

you know -- we see his face constantly.

這可不是什麼好處

So there really was no tolerable margin of error.

正因如此我們不能出錯

There were two studios involved: Warner Brothers and Paramount.

過程中由兩個製片公司一起合作: 華納兄弟和派拉蒙

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

當然，他們都深信這會是一部很讚的電影

but it was a very high-risk proposition.

無庸置疑也是很冒險的計劃

There was lots of money and reputations at stake.

攸關大把金錢與名聲

But we believed that we had a very solid methodology

我們認為我們有很可靠的方法

that might work ...

有可能成功的

But despite our verbal assurances,

但我們的口頭保證不夠

they wanted some proof.

他們要看到證據

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

所以在2004年，他們要求看到班傑明的試鏡畫面

And we did it in about five weeks.

我們做了約五週

But we used lots of cheats and shortcuts.

用了些旁門左道

We basically put something together to get through the meeting.

基本上我們東加西減地混過會談

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

現在放給各位看。這是第一個班傑明巴頓

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

在這裡可以看出是電腦合成出來的頭

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

其實不錯，可以接到演員的身體

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

也算成功，製片那邊終於放下大石

After many years of starts and stops on this project,

經過多年走走停停

and making that tough decision,

作出一些困難的決定

they finally decided to greenlight the movie.

他們終於同意了

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

我還記得，當我接到電話

to say the movie was a go,

恭喜我們電影終於可以開拍

I actually threw up.

我還吐了

(Laughter)

(笑聲)

You know, this is some tough stuff.

這是艱難的工作

So we started to have early team meetings,

我們開始初期的團隊會議

and we got everybody together,

集合大家

and it was really more like therapy in the beginning,

剛開始有點像在諮商治療

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

告訴彼此，我們一定做得到的

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

要能撐過電影一個小時

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

而這不是特效片，必須用真人

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

這真的感覺像是戒酒那種12步計畫

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

當然第一步就是: 承認問題的存在

So we had a big problem:

我們問題可大了

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

不曉得到底怎麼開始

But we did know one thing.

但我們知道

Being from the visual effects industry,

作視覺特效這行

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

我們和大衛相信的，只有足夠時間

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

足夠資源，天啊，希望我們有足夠資金

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

我們也有足夠的熱誠，成功的決心

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

當你面對這樣的情況

of course you've got to break it down.

當然得先開始分析問題

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

你面前有個大問題，把它拆成小問題

and you start to attack that.

個個攻破

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

主要集中在三個點

We needed to make Brad look a lot older --

我們要布萊德看起來很老

needed to age him 45 years or so.

我們要幫他加個45歲左右

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

我們得確定它有布萊德的臉部特徵

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

臉部肌肉的抽動，各種他特有的細微變化

and have that translate through our process

經過我們的製作過程

so that it appears in Benjamin on the screen.

讓班傑明能夠呈現於大螢幕

And we also needed to create a character

我們也需要創造一個

that could hold up under, really, all conditions.

能夠做出每件事的角色

He needed to be able to walk in broad daylight,

他要能走在陽光下、

at nighttime, under candlelight,

夜晚、燭光下

he had to hold an extreme close-up,

近拍不能露餡

he had to deliver dialogue,

要能唸台詞

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

要能跑、能流汗

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

能洗澡、能哭

he even had to throw up.

甚至要能嘔吐

Not all at the same time --

當然不是同時

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

但他要能做到每件事

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

而且要能撐住電影前一小時

We did about 325 shots.

我們拍了325個景

So we needed a system that would allow Benjamin

我們需要一個系統

to do everything a human being can do.

讓班傑明做出人類能做的所有事

And we realized that there was a giant chasm

我們知道2004年當時最先進的技術

between the state of the art of technology in 2004

和我們所想要的技術

and where we needed it to be.

仍有很大一段距離

So we focused on motion capture.

所以我們專注於動態捕捉技術

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

我相信大部分的人都看過動態捕捉

The state of the art at the time

當時最先進的技術是

was something called marker-based motion capture.

利用標記點的動態捕捉

I'll give you an example here.

給大家看個例子

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

主要就是，穿著緊身衣

and they put some reflective markers on your body,

在身上標出一些反光標記點

and instead of using cameras,

物體四周放置紅外線感應

there're infrared sensors around a volume,

而不是攝影機

and those infrared sensors track the three-dimensional position

這些紅外線感應，會追蹤標記點

of those markers in real time.

當時的3D位置

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

接著動畫師把標記點的數據

and apply them to a computer-generated character.

輸入到電腦做出來的角色上

You can see the computer characters on the right

可以看到右邊的動畫角色

are having the same complex motion as the dancers.

也能做出和舞者一樣的複雜動作

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

我們也看了一些

that were using facial marker tracking,

使用臉部標記點的電影

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

就是把點標記在臉上

and doing the same process.

過程相同

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

可以看到，感覺蠻假的

That's not terribly compelling.

這樣可沒法說服別人

And what we realized

我們了解到

was that what we needed

我們所需要的

was the information that was going on between the markers.

是點與點之間的數據

We needed the subtleties of the skin.

我們要皮膚的微妙變化

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

要能看到骨頭到肌肉到皮膚的變化

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

需要皺紋、酒窩、細紋等等

Our first revelation was to completely abort and walk away from

我們了解現在需要完全放棄

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

當時最先進的技術

So we aborted using motion capture.

我們放棄使用動態捕捉

And we were now well out of our comfort zone,

丟下了自己所熟悉的技術

and in uncharted territory.

到了陌生的領域

So we were left with this idea

我們有了一種想法

that we ended up calling "technology stew."

後來我們稱為科技大雜燴

We started to look out in other fields.

我們開始往別的領域找

The idea was that we were going to find

目的是要尋找出

nuggets or gems of technology

其他科技領域像是醫學影像

that come from other industries like medical imaging,

或電玩空間技術

the video game space,

之中的稀世珍品

and re-appropriate them.

拿來重新研究

And we had to create kind of a sauce.

雜燴也需要點醬料

And the sauce was code in software

就是軟體程式碼

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

讓這些不同的領域的技術

to come together and work as one.

能整合在一起

Initially, we came across some remarkable research

起先，我們找到一項傑出的研究

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

是70年代初期，Paul Ekman博士提出的

He believed that he could, in fact,

他相信他有辦法

catalog the human face.

將人臉目錄化

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

他想到使用臉部表情辨識系統(FACS)

He believed that there were 70 basic poses

他相信人的臉型

or shapes of the human face,

有70種基本表情動作

and that those basic poses or shapes of the face

只要把基本動作或臉型

can be combined to create infinite possibilities

結合起來就能創造出

of everything the human face is capable of doing.

人臉做的出的無限表情

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

不論年齡、種族、文化、性別，這70種動作都是一樣的

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

這變成我們往後的基礎點

And then we came across some remarkable technology

之後我們發現了項非凡的技術

called Contour.

叫做輪廓提取(Contour)

And here you can see a subject having phosphorus makeup

現在看到的，是在人臉上

stippled on her face.

輕輕塗上磷

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

接下來是表面捕捉

as opposed to a marker capture.

正好與點的捕捉相反

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

接著站在一排電腦攝影機前面

and those cameras can, frame-by-frame,

攝影機就會一景一景的

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

重製出人當時的動作

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

立刻就能得到3D數據

And if you look in a comparison,

各位可以看看比較

on the left, we see what volumetric data gives us

左邊是體積數據的結果

and on the right you see what markers give us.

右邊是標記點的

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

可以清楚看出，我們有很大的進步

But these were the early days of this technology,

但這還是初期的結果

and it wasn't really proven yet.

還沒完全成功

We measure complexity and fidelity of data

但我們依據多邊形數量來計算

in terms of polygonal count.

數據的複雜性和準確性

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

左邊可以看到有十萬的多邊形

We could go up into the millions of polygons.

還可以做到幾十萬個

It seemed to be infinite.

甚至無限多

This was when we had our "Aha!"

這就是我們"啊哈"的時候

This was the breakthrough.

這是個大突破

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

這時我們才告訴自己

This is actually going to work."

一切都會沒事

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

"啊哈"就是，我們或許可以請布萊德彼特

and we could put Brad in this device,

使用這個裝置

and use this Contour process,

再搭配輪廓提取

and we could stipple on this phosphorescent makeup

塗上發磷光的顏料

and put him under the black lights,

用螢光燈照射

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

拍攝Ekman的FACS臉部動作時

performing Ekman's FACS poses.

就能即時掃描出來

Right? So, effectively,

對吧，所以最後

we ended up with a 3D database

我們有了所有布萊德彼特的臉

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

能做的所有表情

(Laughter)

(笑聲)

From there, we actually carved up those faces

之後，將他的這些臉

into smaller pieces and components of his face.

劃分成小塊小塊的

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

最後有了成千上萬的形狀

a complete database of all possibilities

就是這張臉可以做的

that his face is capable of doing.

所有表情

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

看來很不錯，但我們只有44歲的他

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

現在我們需要的是40年後的他

We brought in Rick Baker,

我們找來了Rick Baker

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

Rick是我們這行的化妝、特效

of our industry.

大師之一

And we also brought in a gentleman named Kazu Tsuji,

我們也請來一位Kazu Tsuji先生

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

Kazu Tsuji是位很厲害的超現實雕刻家

And we commissioned them to make a maquette,

我們請他做班傑明的雕塑模型

or a bust, of Benjamin.

也可以說半身像

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

本著盛大揭幕的精神

I had to unveil something.

我一定要揭露一下

So this is Ben 80.

這是80歲的班傑明

We created three of these:

我們製作了三個

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

80歲、70歲、60歲

And this really became the template for moving forward.

這就是之後我們依照的模板

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

這是以布萊德的模樣刻的

So, in fact, anatomically, it is correct.

所以當然是正確的樣貌

The eyes, the jaw, the teeth:

眼睛、下巴、牙齒...

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

每一樣都是真人比例

We have these maquettes scanned into the computer

我們把雕塑圖掃瞄到電腦

at very high resolution --

使用超高解析

enormous polygonal count.

很多的多邊形

And so now we had three age increments of Benjamin

現在電腦裡有了三種班傑明

in the computer.

年老的樣子

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

但我們需要更多的數據

We went through this process, then, called retargeting.

接下來的過程，叫做重定向

This is Brad doing one of the Ekman FACS poses.

這是布萊德做的Ekman FACS

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

這是獲得的數據

the model that comes from that.

弄出來的樣子

Retargeting is the process of transposing that data

重定向就是將這些數據轉移到

onto another model.

另一個模型上

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

利用依照布萊德做出的

was made from Brad,

班傑明的半身像

we could transpose the data of Brad at 44

我們就能把44歲布萊德的數據轉移

onto Brad at 87.

到87歲的布萊德

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

所以我們有了布萊德彼特

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

87、70、60歲的所有臉部3D資料

Next we had to go into the shooting process.

接下來就是進行拍攝了

So while all that's going on,

同時間

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

我們去了紐奧良和其他各地

And we shot our body actors,

我們先拍了身體演員

and we shot them wearing blue hoods.

讓他們穿了藍色頭套

So these are the gentleman who played Benjamin.

這位先生飾演班傑明

And the blue hoods helped us with two things:

藍色頭套的功用是

one, we could easily erase their heads;

第一，可以輕易消除他們的頭

and we also put tracking markers on their heads

還有，可以直接標上追蹤點

so we could recreate the camera motion

以便讓我們能在片廠重製

and the lens optics from the set.

攝影動作和鏡片光學

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

現在需要拍布萊德的表演，來結合虛擬的班傑明

And so we edited the footage that was shot on location

我們重新編輯了實體拍攝的

with the rest of the cast and the body actors

其他演員片段

and about six months later

約半年後

we brought Brad onto a sound stage in Los Angeles

布萊德來到了洛杉磯片廠

and he watched on the screen.

他看了已經拍好的畫面

His job, then, was to become Benjamin.

他的工作就是變成班傑明

And so we looped the scenes.

我們重複放了片段

He watched again and again.

他看了又看

We encouraged him to improvise.

我們鼓勵他即興唸台詞

And he took Benjamin into interesting and unusual places

他也把班傑明演的有趣獨特

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

出乎我們料想

We shot him with four HD cameras

我們用四台高清攝影機拍攝

so we'd get multiple views of him

為了取得各種角度的畫面

and then David would choose the take of Brad being Benjamin

然後大衛會選最符合該片段

that he thought best matched the footage

的臉部角度

with the rest of the cast.

以搭配身體部分

From there we went into a process called image analysis.

再來的過程是影像分析

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

可以看到，選好的片段

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

資料轉換成班傑明87歲

And so, what's interesting about this is

最有趣的就是

we used something called image analysis,

我們使用畫面分析

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

擷取各時間點班傑明臉部的每個部分

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

所以我們可以隨便選，例如他的左眉

And the software would tell us that, well,

軟體就會告訴我們

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

在第14景，左眉從這裡移到這裡

and it concludes moving in frame 32.

然後在第32景停止

And so we could choose numbers of positions on the face

可以隨便取得任何一個表情的

to pull that data from.

臉部數據

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

再來就是剛剛提過，科技大雜燴的醬料

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

獨門配方就是，一個軟體能讓我們

match the performance footage of Brad

用布萊德的表演

in live action with our database of aged Benjamin,

配上電腦裡班傑明年老的

the FACS shapes that we had.

FACS數據

On a frame-by-frame basis,

有了各景的基礎

we could actually reconstruct a 3D head

我們可以重建出完全符合

that exactly matched the performance of Brad.

布萊德演出的3D頭

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

這就是電影最終畫面

And here you can see the body actor.

可以看到身體演員

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

這個我們稱呼為死人頭，跟Jerry Garcia無關

And then here's the reconstructed performance

這個是重建的演出

now with the timings of the performance.

配上正確的時間點

And then, again, the final shot.

最後，完成的畫面

It was a long process.

是很冗長的過程

(Applause)

(掌聲)

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

下一段我只能稍微帶過

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

要細說可能需要再做一次TEDTalk

We had to create a lighting system.

我們必須創造出燈光系統

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

這部分最難的是，創造班傑明每一場景

for every single location that Benjamin had to appear

的各種燈光環境

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

到時候把頭接上才不會怪怪的

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

而且才能完美搭配真實演員的

in the real world.

身體演出部份

We also had to create an eye system.

我們也需要創造眼睛系統

We found the old adage, you know,

因為俗話說的對

"The eyes are the window to the soul,"

"眼睛是靈魂之窗"

absolutely true.

真是一點也沒錯

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

關鍵就是，要讓大家能在班傑明眼中

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

感覺的到溫暖，感覺的到人性

and feel his intent coming through the eyes,

從眼中感受的到熱誠

then we would succeed.

這樣就是成功了

So we had one person focused on the eye system

我們有個同仁專門弄眼睛系統

for almost two full years.

快兩年的時間

We also had to create a mouth system.

再來是嘴巴系統

We worked from dental molds of Brad.

我們從布萊德的齒模下手

We had to age the teeth over time.

再來弄不同年紀的牙齒

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

還有要個能夠清晰發音的舌頭

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

有個完整程式負責清晰地發音

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

有位同仁花九個月的時間，致力於舌頭

He was very popular.

他很受歡迎

Skin displacement: another big deal.

皮膚替換: 另一個挑戰

The skin had to be absolutely accurate.

皮膚必須完全精確

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

他待過安養院、療養院

around other old people,

身邊有其他老人

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

所以他必須看起來跟其他人一樣

So, lots of work on skin deformation,

老化皮膚是項大工程

you can see in some of these cases it works,

可以看到有時候成功

in some cases it looks bad.

有時候失敗

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

這是我們非常早期的成果

So, effectively we created a digital puppet

蠻成功的做出數位人偶

that Brad Pitt could operate with his own face.

讓布萊德彼特可以自己控制

There were no animators necessary to come in and interpret behavior

不需要動畫師來製作

or enhance his performance.

或加強表演

There was something that we encountered, though,

我們還經歷了一個過程

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

我們稱作是數位肉毒桿菌

So, as things went through this process,

在過程中

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

芬奇說，這修飾了最後演出

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

製作過程與技術唯一做不到的

is they couldn't understand intent,

是無法理解意圖

the intent of the actor.

演員的意圖

So it sees a smile as a smile.

微笑就是微笑

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

它不會分辨是諷刺的笑、或開心的笑

or a frustrated smile.

或挫折的笑

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

沒辦法理解其中的情緒意思

But we ended up calling the entire process

我們最後稱呼這個過程

and all the technology "emotion capture,"

情緒捕捉技術

as opposed to just motion capture.

與動態捕捉相反

Take another look.

再看一次

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

布萊德彼特: 我聽到媽和Tizzy講悄悄話

and they said I was gonna die soon,

說我活不了多久

but ... maybe not.

但是... 也許他們說錯了

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

這就是18分鐘內解釋完

(Applause)

(掌聲)

A couple of quick factoids;

再補件有趣的事

it really took 155 people over two years,

整個過程大概花了155人兩年的時間

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

我們甚至沒聊到60個髮型和數位剪髮

But, that is Benjamin. Thank you.

但，這就是大家的班傑明，謝謝