ADM-201 dump PMP dumps pdf SSCP exam materials CBAP exam sample questions

撞脸是什么体验? – 译学馆
未登录,请登录后再发表信息
最新评论 (0)
播放视频

撞脸是什么体验?

Does Someone Else Have Your Face? Doppelgängers Explained!

PBS数字工作室
Thank you for Brilliant.ORG for supporting PBS digital studios.
感谢Brilliant网站对PBS数字工作室的支持
Oh, hey smart people.
嘿 聪明人
I was just, you know, selfie-ing.
我刚刚只是在
On the Google Arts and Culture app.
用谷歌艺术文化应用自拍
It finds your face in famous artwork,
你会在艺术品上发现你的脸
your art doppelganger if you will.
如同你的艺术品分身
Mine is… um, I’m not sure.
我的 好像不太像
But this made me wonder
但这使我想知道
why do some faces look the same to us?
为什么有些面孔在我们看来一样?
And how do we even recognize faces?
我们如何识别人脸呢?
I mean, there’s nearly 8 billion people on earth.
我的意思是 在地球上有近乎80亿的人
Does someone else have your face?
还有其他人和你长得像吗?
《聪明刷》
Everyday as you scroll through social media,
当你每天浏览社交媒体时
you see dozens, maybe hundreds of faces.
你会看见几十甚至上百个的面孔
Your brain analyzes them, and matches them to an identity,
你的大脑会分析并和某人进行匹配
without you even consciously thinking about it.
甚至你都没意识到
But what would you do if you were scrolling
但是 如果你在浏览Instagram时
through Instagram and saw your own face?
看到自己的脸 你会做什么呢?
This is Amanda Green.
这是阿曼达·格林
This is… not Amanda Green.
这不是阿曼达·格林
And this IS Amanda Green again.
这又是阿曼达·格林
I’ll let her explain.
我让阿曼达解释一下
A couple of years ago I found my look-alike,
几年前我发现和我长得像的人
and it was the strangest but most exciting thing
这可能是发生在我身上的
that’s probably ever happened to me.
最奇怪也是最有趣的事
And now I have a great story to tell at parties.
这给了我在派对上的谈资
A friend of Amanda’s saw a picture online
阿曼达的朋友在网上看到
that looked like her.
一张像她的照片
But it wasn’t Amanda.
但并不是阿曼达
It was someone who shared Amanda’s face.
只是有人长得和阿曼达很像
After she posted the photos, the internet found her doppelganger.
她上传照片后 互联网发现了她的“替身”
A girl named Meredith from Indiana,
来自印第安纳州的一个叫做梅瑞狄斯的女孩
who-WOW-looks exactly like her.
确实和她长得很像
At least I think so.
至少我这么觉得
I think it’s when it was
我想当它在
trending in second place on Buzzfeed,
Buzzfeed上的热门话题排名第二
like, under Kim Kardashian
在金·卡戴珊之后
that I was like this is really bizarre.
我觉得这太奇怪了
But after Amanda and Meredith’s photos went viral,
但在阿曼达和梅瑞狄斯的照片火起来之后
some people argued that they didn’t actually look alike,
一些人认为她们并不是真的长得像
because it’s the internet,
因为这是互联网
and people will argue about anything.
人们会为任何事争论
Some people were like “You guys look absolutely nothing alike.”
一些人会说 你们长得一点也不像
I don’t get why this even a story
我不明白为什么这都能成新闻
and other people were like
但是其他人会说
“Okay, that’s the same person who posted twice. “
只是同一个人发了两次照片而已
It was like Yanny and Laurel.
就像和雅尼和劳雷尔一样
But for faces, even Amanda’s family was divided.
但是对于长相 即使阿曼达的家人也意见不一
My mom did think we looked alike.
我妈妈觉得我们长得像
My husband, he said he couldn’t really see it,
我丈夫却说他真的看不出哪里像
but he kind of sucks at faces anyways.
但不管怎么说他不太懂得分辨脸
So why do people see different things
那么为什么人们会在同样两张脸上
in the same two faces?
看出不同的东西呢?
Well the answer gives us a clue about what doppelgangers really are:
这个回答给了我们一个关于“替身”到底是什么的线索
A strange side effect of how our brains process faces.
即大脑处理面部信息的一个奇怪的副作用
So what is a face?
所以到底什么是面孔?
Well “Duh, Joe.
呃 每个人
It’s the eyes and the mouth and the nose,
都有一双眼睛 一张嘴 和一个鼻子
and the dimples in the cheeks
还有脸颊上的酒窝
and the color of the eyebrows.”
以及眉毛的颜色
All these features together create a picture
这些所有的特征一起构成一幅画面
that you present to the world, your face.
就是你呈现在世界上的画面——你的脸
But a picture of a face isn’t really
但一张脸的图像并不是
what your brain sees when you look at someone.
你真正看到某人时脑海里产生的画面
When we recognize anything,
当我们在辨别一些东西的时候
we are comparing what we see
我们会将我们看到的
with stored mental picture that’s encoded in our brain.
和脑内经过编码的存储画面进行比较
Turns out special cells in our brains
结果是只有当我们看到人脸时
are active only when we look at faces,
大脑里某些特殊细胞才会变得活跃
and they’re not active when we look at other things.
我们看其他东西时它们并不会活跃
So what are these facial recognition neurons actually seeing?
那么这些面部识别神经元到底看到了什么呢?
This person in Ariana Grande’s “Thank U Next” video
在A妹的Thank U Next这首歌MV中的这个人
really looks like Lindsay Lohan,
真的很像琳赛
but it isn’t Lindsay Lohan.
但她并不是琳赛·罗韩
Sure her red hair and the “Mean Girls” outfit
当然她的红头发还有“坏女孩”装扮
are part of why we’re fooled.
是我们被愚弄的部分原因
But watch what happens when we overlay their faces.
但看看当我们重叠她们的脸会发生什么
The geometry of the face,
脸部的几何图形
how the features are arranged with respect to one another, is almost identical,
这些特征的排列方式几乎是完全相同的
even if the specific features, like the eyebrows and the tip of the nose are different.
即使像眉毛和鼻尖这样的特定特征不同
The same thing is true for Amanda and Meredith.
阿曼达和梅瑞狄斯也是一样的道理
If you really look at our mouth and nose,
如果你仔细观察我们的鼻子和嘴巴
it might not look similar,
可能看起来不像
but they’re in like the exact same place.
但它们完全在同一个地方
Like if you drew lines across your face,
就像你在你的脸上划线
like my nose, my mouth, and my eyes
我的鼻子 嘴巴还有眼睛
hit at the exact same places as hers.
和她在同样的位置
In the areas of our brain that recognize faces,
在我们大脑识别人脸的区域
we think certain neurons wire for eyes that are this far apart
我们认为一些控制眼睛的神经元 隔着这么远
others for a mouth here, or a nose is this long.
而其他负责嘴的在这里 鼻子的是这么长
This combination of nerves wiring creates a map,
这些神经连接的组合创造了一幅地图
or a code, called a facial schema.
或者说是一个密码 也叫面部图式
To our brains, a face isn’t a picture, it’s a pattern.
对于大脑来说 脸不是图片 是一个模式
And if one of those coded patterns shows up somewhere unexpected,
如果其中一种编码模式出现在意料之外的地方
we can see a face even if it’s not there
我们就会看到一张并不存在的脸
. It’s an effect called Pareidolia.
这是一种叫做幻想性错觉的效应
Take a look at this photo of George W Bush and Dick Cheney.
看看这张照片里的小布什和迪克
It’s actually George W Bush… and George W Bush.
实际上都是小布什本人
Did you notice?
你注意到了吗?
Recognize this guy?
认识这个人吗?
Maybe he seems a little familiar,
可能他看起来有点眼熟
but you can’t quite place it.
但是你不能很好的认出来
It’s two people – Harrison Ford and George Clooney.
这是两个人——哈里森福特和乔治克鲁尼
When they’re together, our brains decode them as just one face.
当把他们合在一起的时 我们的大脑将他们解码为一张脸
And here are two photos of the same person.
这是同一个人的两张照片
Give them a quick look.
快速地浏览一下
Okay, let’s flip the image.
现在让我们翻转图像
All of these are examples of how we get fooled,
这些都是我们被捉弄的例子
because our brains usually don’t pay attention
因为我们的大脑通常对脸部的细节
into the details of faces.
不太在意
We don’t have cameras in our heads.
我们大脑中没有相机
We’re running pattern recognition software between our ears.
我们在两耳之间运行模式识别软件
And I guess this means we need a firmware upgrade.
我想这意味着我们需要升级固件
Of course, the more time we spend with people,
当然 我们与人相处的时间越多
the better we can tell their face apart from others,
我们越能区分他们的脸和其他人的脸
because our face software gets help from our long term memory.
因为我们的脸部识别软件是从长时记忆中发展而来
Like, we could all tell if the President was replaced
如果总统被一个叫Dave的替身或别的谁代替
by somebody dopple named Dave or something, right?
我们都可以分辨出来 对吧?
You’re a very handsome man. Thank you Mr. President.
您真是个帅气的人 谢谢您尊敬的总统
You come for the science,
你本是为了科学而研究
but you stay for the very current pop-culture references.
但却聚焦在了这些流行文化的佐证上
But if you don’t know two people,
如果你不认识这两个人
like Meredith and Amanda
比如说梅瑞狄斯和阿曼达
you’re more likely to think they look alike.
你更有可能认为她们长得像
Researchers at Cambridge University developed
剑桥大学的研究人员开发了一种
a face memory test that you can take for yourself.
可以自己做的面部记忆测试
As it turns out, I am very good at remembering new faces.
事实证明 我很擅长记忆新面孔
And speaking of remembering faces,
说到记忆面孔
did you notice we just switched Amanda and Meredith’s photos?
你注意到我们把阿曼达和梅瑞狄斯的照片换了吗?
Some people are just better at facial recognition than others,
有些人就是比其他人更擅长面孔识别
and scientists don’t really know why.
科学家也不知道原因
People who don’t think Amanda and Meredith look alike
那些不认为阿曼达和梅瑞狄斯长得像的人
are probably able to sense details
可能能够在他们的面部识别模式中
in their face patterns that others can’t.
感知到别人无法感知的细节
But what if someone’s life depended on
但如果某人的生命取决于
your ability to recognize and tell faces apart?
你识别和区分人脸的能力呢?
Would you trust yourself?
你会相信你自己吗?
That’s exactly what Dr. Teghan Lucas does.
这正是卢卡斯博士所做的
She is a professional face identifier.
她是专业的面部识别专家
Or more accurately, a forensic anthropologist
更准确的说 是一名法医
who specializes in facial anatomy.
专精面部解剖学
If police need to identify someone from an image,
如果警察需要辨认图像中的人
they turn to her for help.
他们会向她求助
To the naked eye,
肉眼来看
these people may seem to look the same,
这些人都可能看起来一样
and then if you call someone like me in,
但如果你让像我一样的人看
we would be able to tell slight differences between them,
我们可以看出他们之间的细微差别
because we’re trained to look for very, very miniscule things.
因为我们被训练去寻找非常非常小的东西
Let’s imagine a face as a Rubik’s cube.
让我们把脸想象成一个魔方
It might seem like you can only get so many
在你得到一份复写之前
combinations of nose, eye, and mouth shapes
似乎你只能得到许多
before you get a duplicate.
鼻子 眼睛和嘴的形状的结合体
That’s something that my researchers and I talk about:
这是我和我的研究人员讨论的问题
probabilities of facial characteristics and body characteristics.
面部特征和身体特征的概率
You know, the probability of finding
也就是发现
two people with the same face
两个人拥有同样的脸
or two people with the same body.
或者两个人身材一样的概率
And we found that these probabilities are
我们发现这些概率
comparable with DNA and fingerprints.
可与DNA和指纹匹配概率相对应
So our face is just as unique as a fingerprint or as DNA.
我们的脸就像指纹和基因一样独一无二
If we factor in just 8 facial measurements,
如果我们考虑八种面部测量因素
the odds of two people having the same face
两人拥有同样相貌的几率
are about 1 in a trillion.
是一万亿分之一
So, basically impossible.
因此 基本上是不可能的
If I look at a nose, and I see maybe wide or skinny,
如果我看一个人的鼻子 我看的是鼻子宽还是窄
or how much it sticks up.
或者鼻梁有多高
Dr Lucas sees nose width, bridge width,
卢卡斯博士看的是鼻子和鼻梁的宽度
tip size, tip shape, tip angle,
鼻尖的大小 鼻尖的形状
angle in relation to the chin, the forehead,
和鼻尖与下巴 前额的角度关系
how much a nose protrudes, she knows her stuff.
鼻子突出了多长 她知道该看哪里
Even if you take something as simple as the ear,
即使你只拿像耳朵这样简单的器官举例
the ear is unique between each and every individual.
每个人的耳朵也是独一无二的
And there’s enough anatomical characteristics on the ear
并且耳朵有很多结构上的特点
that we can actually identify someone from the ear.
我们可以通过耳朵辨别出一个人
And I’ve had cases of that.
并且我已有这样的案例
Someone has robbed a bank with a balaclava
有人错误地戴上巴拉克拉法帽
that was on completely the wrong way,
抢劫了一个银行
and his ear was showing, and we catched them up.
他的耳朵露了出来 我们抓住了他们
And you know I asked her about
你知道我一定会问她关于
the David Schwimmer look-alike thief.
有关与大卫·休默长得像的盗贼的事
Yeah, I picked quite a few differences,
是的 我挑选出一些不同之处
mostly in the nose and he had very square chin,
大部分在鼻子上 他的下巴非常方正
the David Schwimmer look alike.
和大卫·休默非常像
And when I looked at it, it was sort of like oh yeah
当我看着他的时候 他有点像 哦 是的
I could put this to rest in 5 minutes.
我需要一些时间来辨别
So I asked her to compare Amanda and Meredith.
所以我请求她对比阿曼达和梅瑞狄斯
Here’s what she found
这是她的发现
So Amanda had a triangular eyebrow shape
阿曼达的眉毛是三角形的
meaning she had a bit of an arch in the
意味着她眉毛最上方的
most superior corner of the eyebrow
眉毛最上角是拱形的
Whereas Meredith had a straight eye brow.
然而梅瑞狄斯是直眉
Amanda had an oval face shape.
阿曼达是卵形的脸
Whereas Meredith had a more elliptical face shape.
然而梅瑞狄斯的脸型更加的椭圆
Amanda had a V shaped upper lip.
阿曼达的上唇是V型的
But in Meredith, she has what we call a cupid’s bow.
但是梅瑞狄斯有我们所说的丘比特之弓的形状
Meaning, the project up with are more rounded
意味着上唇要更圆一些
And Amanda had a rounded bottom lip.
阿曼达的下嘴唇是圆形的
Whereas Meredith had a W shaped bottom lip.
然而梅瑞狄斯的下嘴唇是W形的
So it kind of had those points.
因此 她们某些部分是相似的
But, on the bottom,
但在下面的部分
in Amanda, she had what we call a nasolabialis fold.
对于阿曼达来说 她有我们所说的鼻唇褶皱
It’s these folds here.
这些褶皱在这里
And she’s got them quite prominently even when she doesn’t smile.
即使她不笑 也表现得很明显
Whereas Meredith doesn’t have those folds at all, unlike Amanda.
然而梅瑞狄斯完全没有这些褶皱 不像阿曼达
Doppelgangers are never identical matches,
面貌相似的人从来不是完全一样的
but they don’t have to be to cause problems.
但他们不一定会造成问题
Dr.Lucas is working on a case right now
卢卡斯博士现在正在处理一个
involving someone who’s been in prison for over a decade,
牵涉到一个坐了十多年牢的人的案子
maybe by mistake,
也许判错了
because their face might look enough like someone else’s.
因为他们的脸看起来很像别人的脸
Facial recognition is very unreliable.
面部识别是很不可靠的
And in that moment,
在那一刻
when you’re actually calling upon someone to recall their memory,
当你要求某人去回忆时
entirely unreliable.
是完全不可靠的
So if we get fooled so often,
因此 如果我们常常被愚弄
maybe computers could do better than humans?
电脑可能比我们做得更好吗?
But I don’t know.
但是 我不知道
Computers can see a face in someone’s knee.
电脑能够在某人的膝盖上看到一张脸
Or there was the time that an automated surveillance system in China
或者曾经有一个中国的自动监控系统
publicly shamed a woman for jaywalking after
在公交车上捕捉到一个女士的脸
after seeing her face on the side of a bus.
却通报她横穿马路
But what if we let computers teach themselves how to recognize faces?
但是 如果我们让计算机自学如何识别人脸呢?
Maybe they could get fooled less than we do.
也许它们没我们这么容易被骗
Why don’t you try to find the matching photo?
为什么你不试试着寻找匹配的照片呢?
A computer can do this almost instantly.
计算机几乎可以立即做到
A company called Thorn developed this software
一个叫托恩的公司开发了这个软件
to help find missing kids.
来帮助寻找失踪的孩子
We’ve been able to help investigators identify about 8 kids every day.
我们每天都能帮助调查人员确认8个孩子的身份
Thorn fed images into a machine intelligence,
托恩将图像输入机器智能
and let it learn over time,
并且让它不断的学习
until it was able to identify where a face is,
直到它能够识别出人脸的位置
pick out its important features,
找出它的重要特征
and match similar faces together.
并将相似的脸匹配在一起
What I find most interesting is
我发现最有趣的事情是
that they didn’t train it to see faces like we do.
他们没有训练它像我们一样的辨别面孔
It’s evolved its own way of recognizing faces,
它进化出自己的方式去识别面孔
just like our own brains evolved neural networks
就像我们的大脑进化出神经网络
that help us recognize each other.
来帮助我们去认识彼此一样
But where computers or machine intelligence
但是 电脑或机器智能
really makes a difference is that it’s capable of
真正不同的是 它有能力
memorizing such a large number of people.
记住大量的人
I think a human being can remember at most,
我认为一个人 最多只能记住
let’s say 5,000 faces.
假设有5000张脸
For a computer it’s pretty straight forward,
对于电脑来说 这太简单了
with the help of machine intelligence to even memorize
借助机器智能 甚至能够记忆
a million people, or several million people.
一百万人或者几百万人的脸
The AI isn’t perfect, but neither are our own abilities.
虽然人工智能并不完美 但我们自己的能力也不完美
But our powers combined can have serious potential.
但我们的力量结合起来能够有无穷的潜力
If we think of a face like a lost needle,
如果我们把脸想象成一根丢失的针
AI makes the haystack way smaller from the get-go.
人工智能在一开始就缩小范围
Yeah and I think that’s really where we’re mostly going towards.
我认为这才是我们的目的所在
Like 5 million faces, where is this one particular person.
比如五百万张脸中 这个特别的人在哪里
And enable us to do things we’re unable to do ever before.
让我们能够做以前无法做的事情
That’s very exciting.
这是非常令人兴奋的
So, science tells us that technically doppelgangers don’t exist.
科学告诉我们 严格来说 面部相似的人是不存在的
Every face that’s ever been is unique
每张脸都是独一无二的
yours, mine
不管是你的还是我的
even between identical twins.
甚至在双胞胎之间
But science has also shown us
但是科学也同时展示给我们
that we’re really good at fooling ourselves.
我们很擅长欺骗我们自己
Even when something is right in front of our face.
即使有些事情就发生在我们面前
Stay curious.
保持一颗好奇心
We want to say a big thank you for Brilliant.ORG
我们非常感谢Brilliant网站
to support PBS Digital Studios and our show.
支持PBS数字工作室和我们的节目
Is it me? or is it some evil look alike?
这是我吗 还是什么长得很像的魔鬼?
I’ll never tell.
我也说不上来
But i do know so much easier to learn
但是我知道如果我有一个“替身”的话
all the things if i had a doppelganger running around.
学习所有的东西会容易得多
But then it has figured out how we could all share brain
但它后来发现了我们如何共享大脑
and all that how much easier way to sharpen your scientific skills
所有这些都是提高你技能的简单方法
and have fun doing it.
祝你玩的开心
Please check out Brilliant.ORG
请看看Brilliant网站
Brilliant is about learning actual problem solving not just watching stuff.
Brilliant是学习解决实际问题的 而不只是观看视频
They have sequences focus on machine learning and neutral netwoks.
他们把重点放在机器学习和神经网络上
We sure exactly what people like Ruben in his team used to train their computers to do.
我们完全可以确定像鲁本这样的人 过去是如何训练他们的电脑的
The universe’s most complex computers: our brains already do so well.
宇宙中最复杂的电脑:我们的大脑已经做得很好了
To solve problems like this,
为了解决这样的问题
And ones we haven’t even thought of yet.
还有一些我们还没想到的问题
You need their free work for thinking about and solving it.
你需要他们的自由工作来思考和解决它
Brilliant is a place that you can do that.
Brilliant是一个可以做到这些的地方
Brilliant: math and science done right.
Brilliant:擅长数学与科学
Proud to support it’s okay to be smart.
为你们支持《聪明刷》感到自豪
To learn more about Brilliant.
了解更多关于Brilliant
You can go to Brilliant.ORG to slash be smart.
你可以登录Brilliant.org/besmart

发表评论

译制信息
视频概述

真正面部相似的脸是不存在的,只是我们难以发现区别罢了,在人工智能面前,就连双胞胎都能发现区别之处

听录译者

收集自网络

翻译译者

cure-allྀི

审核员

审核员XY

视频来源

https://www.youtube.com/watch?v=4AQm22YQfYU

相关推荐