返回首页

FeiFeiLi_2015-_我们怎么教计算机理解图片？_
Let me show you something. 我先来给你们看点东西。
(Video) Girl: Okay, that's a cat sitting in a bed. （视频）女孩:好吧,这是只猫,坐在床上。
The boy is petting the elephant. 一个男孩摸着一头大象。
petting:n.亲吻抚摸;调情;v.抚摸;亲吻;调情;爱抚;(pet的现在分词)
Those are people that are going on an airplane . 那些人正准备登机。
airplane:n.飞机;
That's a big airplane. 那是架大飞机。
Fei-Fei Li: This is a three-year-old child describing what she sees in a series of photos. 李飞飞:这是一个三岁的小孩 在讲述她从一系列照片里看到的东西。
describing:v.描述;形容;把…称为;做…运动;(describe的现在分词) series:n.系列,连续;[电]串联;级数;丛书;
She might still have a lot to learn about this world, but she's already an expert at one very important task: to make sense of what she sees. 对这个世界,她也许还有很多要学的东西, 但在一个重要的任务上,她已经是专家了: 去理解她所看到的东西。
make sense of:搞清…的意思;
Our society is more technologically advanced than ever. 我们的社会已经在科技上取得了前所未有的进步。
technologically:adv.科技地;技术上地; advanced:adj.先进的; v.前进; (advance的过去式和过去分词形式)
We send people to the moon, we make phones that talk to us or customize radio stations that can play only music we like. 我们把人送上月球,我们制造出可以与我们对话的手机, 或者订制一个音乐电台,播放的全是我们喜欢的音乐。
customize:vt.定做,按客户具体要求制造;
Yet, our most advanced machines and computers still struggle at this task. 然而,哪怕是我们最先进的机器和电脑 也会在这个问题上犯难。
So I'm here today to give you a progress report on the latest advances in our research in computer vision , one of the most frontier and potentially revolutionary technologies in computer science . 所以今天我在这里,向大家做个进度汇报: 关于我们在计算机视觉方面最新的研究进展。 这是计算机科学领域最前沿的、 具有革命性潜力的科技。
vision:n.视力;美景;幻象;想象力;v.想象;显现;梦见; potentially:adv.可能地,潜在地; revolutionary:adj.革命性的;革命的;彻底变革的;n.改革者;革命者; technologies:n.技术;科技(technology的复数); computer science:n.计算机科学;
Yes, we have prototyped cars that can drive by themselves, but without smart vision, they cannot really tell the difference between a crumpled paper bag on the road, which can be run over, and a rock that size, which should be avoided. 是的,我们现在已经有了具备自动驾驶功能的原型车, 但是如果没有敏锐的视觉,它们就不能真正区分出 地上摆着的是一个压扁的纸袋,可以被轻易压过, 还是一块相同体积的石头,应该避开。
prototyped:n.原型; (prototype的过去分词) crumpled:adj.摺皱的,弄皱的;v.弄皱(crumple的过去分词);
We have made fabulous megapixel cameras, but we have not delivered sight to the blind. 我们已经造出了超高清的相机, 但我们仍然无法把这些画面传递给盲人。
fabulous:adj.难以置信的;传说的,寓言中的;极好的; megapixel:n.兆象素;
Drones can fly over massive land, but don't have enough vision technology to help us to track the changes of the rainforests . 我们的无人机可以飞跃广阔的土地, 却没有足够的视觉技术 去帮我们追踪热带雨林的变化。
Drones:v.嗡嗡叫;嗡嗡响;(drone的第三人称单数) massive:adj.大量的;巨大的,厚重的;魁伟的; technology:n.技术;工艺;术语; track:n.小道;足迹;车辙;轨道;v.追踪;跟踪; rainforests:n.(热带)雨林;
Security cameras are everywhere, but they do not alert us when a child is drowning in a swimming pool . 安全摄像头到处都是, 但当有孩子在泳池里溺水时它们无法向我们报警。
alert:n.警报; adj.警觉的; v.向…报警; drowning:v.溺水;淹没;淹死,溺死;浸泡;(drown的现在分词) swimming pool:n.游泳池;游泳场;游泳馆;
Photos and videos are becoming an integral part of global life. 照片和视频,已经成为全人类生活里不可缺少的部分。
integral:adj.积分的;完整的,整体的;必须的;n.积分;部分;完整; global:adj.全球的;总体的;球形的;
They're being generated at a pace that's far beyond what any human, or teams of humans, could hope to view, and you and I are contributing to that at this TED. 它们以极快的速度被创造出来,以至于没有任何人,或者团体, 能够完全浏览这些内容, 而你我正参与其中的这场TED,也为之添砖加瓦。
generated:v.产生;引起;(generate的过去式和过去分词) contributing:v.捐献,捐赠(尤指款或物);捐助;增加;增进;(contribute的现在分词)
Yet our most advanced software is still struggling at understanding and managing this enormous content . 直到现在,我们最先进的软件也依然为之犯难: 该怎么理解和处理这些数量庞大的内容？
enormous:adj.庞大的,巨大的;凶暴的,极恶的; content:n.内容,目录;满足;容量;adj.满意的;vt.使满足;
So in other words, collectively as a society, we're very much blind, because our smartest machines are still blind. 所以换句话说,在作为集体的这个社会里, 我们依然非常茫然, 因为我们最智能的机器依然有视觉上的缺陷。
collectively:adv.共同地,全体地;
'"Why is this so hard?" you may ask. ”为什么这么困难？“你也许会问。
Cameras can take pictures like this one by converting lights into a two-dimensional array of numbers known as pixels , but these are just lifeless numbers. 照相机可以像这样获得照片: 它把采集到的光线转换成二维数字矩阵来存储 ——也就是“像素”, 但这些仍然是死板的数字。
converting:v.(使)转变,转换,转化;可转变为;归附;(convert的现在分词) two-dimensional:adj.二维的;缺乏深度的; array:n.数组,阵列;排列,列阵;大批,一系列;衣服;v.排列,部署;打扮; pixels:n.像素(组成屏幕图像的最小独立元素);(pixel的复数) lifeless:adj.无生命的;死气沉沉的;无趣味的;
They do not carry meaning in themselves. 它们自身并不携带任何意义。
Just like to hear is not the same as to listen, to take pictures is not the same as to see, and by seeing, we really mean understanding. 就像”听到“和”听“完全不同, ”拍照“和”看“也完全不同。 通过“看”,我们实际上是“理解”了这个画面。
In fact, it took Mother Nature 540 million years of hard work to do this task, and much of that effort went into developing the visual processing apparatus of our brains, not the eyes themselves. 事实上,大自然经过了5亿4千万年的努力 才完成了这个工作, 而这努力中更多的部分 是用在进化我们的大脑内用于视觉处理的器官, 而不是眼睛本身。
Mother Nature:n.大自然;自然界; visual:adj.视觉的,视力的;栩栩如生的; processing:v.加工;处理;审核;数据处理;v.列队行进;缓缓前进;(process的现在分词) apparatus:n.装置,设备;仪器;器官;
So vision begins with the eyes, but it truly takes place in the brain. 所以 视觉”从眼睛采集信息开始, 但大脑才是它真正呈现意义的地方。
So for 15 years now, starting from my Ph.D. at Caltech and then leading Stanford's Vision Lab, 所以15年来,从我进入加州理工学院攻读Ph.D. 到后来领导斯坦福大学的视觉实验室,
Caltech:n.加利福尼亚理工学院;
I've been working with my mentors , collaborators and students to teach computers to see. 我一直在和我的导师、合作者和学生们一起 教计算机如何去“看”。
mentors:n.导师,教练(mentor复数); collaborators:n.[劳经]合作者;投敌者(collaborator的复数);
Our research field is called computer vision and machine learning. 我们的研究领域叫做 计算机视觉与机器学习 。
It's part of the general field of artificial intelligence . 这是AI（人工智能）领域的一个分支。
artificial intelligence:n.人工智能;
So ultimately , we want to teach the machines to see just like we do: naming objects, identifying people, inferring 3D geometry of things, understanding relations, emotions , actions and intentions . 最终,我们希望能教会机器像我们一样看见事物: 识别物品、辨别不同的人、推断物体的立体形状、 理解事物的关联、人的情绪、动作和意图。
ultimately:adv.最终;最后;归根结底;终究; identifying:n.识别,标识;标识关系;v.识别;(identify的现在分词) inferring:v.推断;推论;推理;暗示;意指;(infer的现在分词) geometry:n.几何学;几何结构; emotions:n.强烈的感情;激情;情感;(emotion的复数) intentions:n.目的,意向,意图;打算;(intention的复数)
You and I weave together entire stories of people, places and things the moment we lay our gaze on them. 像你我一样,只凝视一个画面一眼 就能理清整个故事中的人物、地点、事件。
weave:v.编,织;(用…)编成;编造(故事等);n.织法;编法;编织式样; gaze:v.凝视;注视;盯着;n.凝视;注视;
The first step towards this goal is to teach a computer to see objects, the building block of the visual world. 实现这一目标的第一步是教计算机看到“对象”（物品）, 这是建造视觉世界的基石。
In its simplest terms, imagine this teaching process as showing the computers some training images of a particular object, let's say cats, and designing a model that learns from these training images. 在这个最简单的任务里,想象一下这个教学过程: 给计算机看一些特定物品的训练图片, 比如说猫, 并让它从这些训练图片中,学习建立出一个模型来。
images:n.印象;声誉;形象;画像;雕像;(image的第三人称单数和复数)
How hard can this be? 这有多难呢？
After all, a cat is just a collection of shapes and colors, and this is what we did in the early days of object modeling. 不管怎么说,一只猫只是一些形状和颜色拼凑起来的图案罢了, 比如这个就是我们最初设计的抽象模型。
early days:初期;为时尚早;前期;
We'd tell the computer algorithm in a mathematical language that a cat has a round face, a chubby body, two pointy ears, and a long tail, and that looked all fine. 我们用数学的语言,告诉计算机这种算法: “猫”有着圆脸、胖身子、 两个尖尖的耳朵,还有一条长尾巴, 这（算法）看上去挺好的。
mathematical:adj.数学的,数学上的;精确的; chubby:adj.圆胖的,丰满的; pointy:adj.尖的;非常尖的;
But what about this cat? 但如果遇到这样的猫呢？
(Laughter) （笑）
It's all curled up. 它整个蜷缩起来了。
curled:adj.卷曲的;鬈发的;v.[纸]卷曲;环绕(curl的过去分词);
Now you have to add another shape and viewpoint to the object model. 现在你不得不加入一些别的形状和视角来描述这个物品模型。
viewpoint:n.观点;角度;看法;
But what if cats are hidden? 但如果猫是藏起来的呢？
what if:如果…怎么办?
What about these silly cats? 再看看这些傻猫呢？
Now you get my point. 你现在知道了吧。
Even something as simple as a household pet can present an infinite number of variations to the object model, and that's just one object. 即使那些事物简单到只是一只家养的宠物, 都可以出呈现出无限种变化的外观模型, 而这还只是“一个”对象的模型。
household:n.家庭;一家人;同住一所(或一套)房子的人;adj.家庭的;家常的;王室的; infinite:adj.无限的,无穷的; n.无限; variations:n.变奏曲,变更;[生物]变种(variation的复数形式);
So about eight years ago, a very simple and profound observation changed my thinking. 所以大概在8年前, 一个非常简单、有冲击力的观察改变了我的想法。
profound:adj.深厚的;意义深远的;渊博的; observation:n.观察;观测;监视;(尤指据所见、所闻、所读而作的)评论;
No one tells a child how to see, especially in the early years. 没有人教过婴儿怎么“看”, 尤其是在他们还很小的时候。
especially:adv.尤其;特别;格外;十分;
They learn this through real-world experiences and examples. 他们是从真实世界的经验和例子中学到这个的。
real-world:adj.现实生活的;工作的;
If you consider a child's eyes as a pair of biological cameras, they take one picture about every 200 milliseconds , the average time an eye movement is made. 如果你把孩子的眼睛 都看作是生物照相机, 那他们每200毫秒就拍一张照。 ——这是眼球转动一次的平均时间。
biological:adj.生物学的;生物的;与生命过程有关的;加酶的;n.[药]生物制品; milliseconds:n.[计量]毫秒(millisecond的复数形式);
So by age three, a child would have seen hundreds of millions of pictures of the real world. 所以到3岁大的时候,一个孩子已经看过了上亿张的真实世界照片。
That's a lot of training examples. 这种“训练照片”的数量是非常大的。
So instead of focusing solely on better and better algorithms, my insight was to give the algorithms the kind of training data that a child was given through experiences in both quantity and quality. 所以,与其孤立地关注于算法的优化、再优化, 我的关注点放在了给算法提供像那样的训练数据 ——那些,婴儿们从经验中获得的质量和数量都极其惊人的训练照片。
solely:adv.单独地,唯一地; insight:n.洞察力;洞悉; quantity:n.量;数量;大量;数额;
Once we know this, we knew we needed to collect a data set that has far more images than we have ever had before, perhaps thousands of times more, and together with Professor Kai Li at Princeton University, we launched the ImageNet project in 2007. 一旦我们知道了这个, 我们就明白自己需要收集的数据集, 必须比我们曾有过的任何数据库都丰富 ——可能要丰富数千倍。 因此,通过与普林斯顿大学的Kai Li教授合作, 我们在2007年发起了ImageNet（图片网络）计划。
Princeton:n.普林斯顿(美国新泽西州中部的自治市镇); launched:v.发射;发起;开展;开始;(launch的过去式和过去分词)
Luckily, we didn't have to mount a camera on our head and wait for many years. 幸运的是,我们不必在自己脑子里装上一台照相机, 然后等它拍很多年。
mount:n.山;坐骑;山峰;衬纸板;v.登上;爬上;攀登;准备;
We went to the Internet, the biggest treasure trove of pictures that humans have ever created. 我们运用了互联网, 这个由人类创造的最大的图片宝库。
treasure trove:n.无主财宝;宝藏;
We downloaded nearly a billion images and used crowdsourcing technology like the Amazon Mechanical Turk platform to help us to label these images. 我们下载了接近10亿张图片 并利用众包技术（利用互联网分配工作、发现创意或解决技术问题）,像“亚马逊土耳其机器人”这样的平台 来帮我们标记这些图片。
crowdsourcing:众包;群众外包;众包模式; Amazon:亚马逊;古希腊女战士; Mechanical:adj.机械的;力学的;呆板的;无意识的;手工操作的; platform:n.平台; v.把…放在台上[放在高处; label:n.标签;标记;谓;唱片公司;v.贴标签于;用标签标明;
At its peak , ImageNet was one of the biggest employers of the Amazon Mechanical Turk workers: together, almost 50,000 workers from 167 countries around the world helped us to clean, sort and label nearly a billion candidate images. 在高峰期时,ImageNet是「亚马逊土耳其机器人」 这个平台上最大的雇主之一: 来自世界上167个国家的接近5万个工作者, 在一起工作 帮我们筛选、排序、标记了接近10亿张备选照片。
peak:n.高峰; v.达到高峰; adj.最高度的; employers:n.雇主;雇用者;(employer的复数)
That was how much effort it took to capture even a fraction of the imagery a child's mind takes in in the early developmental years. 这就是我们为这个计划投入的精力, 去捕捉,一个婴儿可能在他早期发育阶段 获取的”一小部分“图像。
capture:v.俘虏;捕获;攻占;夺得;刻画,描述;n.(被)捕获;(被)俘获 fraction:n.分数;小部分;小数;少量; imagery:n.像;意象;比喻;形象化; developmental:adj.发展的;启发的;
In hindsight , this idea of using big data to train computer algorithms may seem obvious now, but back in 2007, it was not so obvious. 事后我们再来看,这个利用大数据来训练计算机算法的思路, 也许现在看起来很普通, 但回到2007年时,它就不那么寻常了。
hindsight:n.后见之明;枪的照门; obvious:adj.明显的;显著的;平淡无奇的;
We were fairly alone on this journey for quite a while. 我们在这段旅程上孤独地前行了很久。
fairly:adv.相当地;公平地;简直; journey:n.旅行;行程;vi.旅行;
Some very friendly colleagues advised me to do something more useful for my tenure , and we were constantly struggling for research funding . 一些很友善的同事建议我做一些更有用的事来获得终身教职, 而且我们也不断地为项目的研究经费发愁。
colleagues:n.同事;同行(colleague的复数); advised:adj.考虑过的; v.劝告; (advise的过去分词和过去式) tenure:n.任期;占有;vt.授予…终身职位; constantly:adv.不断地;时常地; funding:n.基金;资金;提供资金;v.为…提供资金;拨款给;(fund的现在分词)
Once, I even joked to my graduate students that I would just reopen my dry cleaner's shop to fund ImageNet. 有一次,我甚至对我的研究生学生开玩笑说: 我要重新回去开我的干洗店来赚钱资助ImageNet了。
reopen:v.重新开业;重新处理;恢复;
After all, that's how I funded my college years. ——毕竟,我的大学时光就是靠这个资助的。
funded:adj.提供资金的;v.提供资金;积存;提供资金偿付的本息;(fund的过去式);
So we carried on. 所以我们仍然在继续着。
In 2009, the ImageNet project delivered a database of 15 million images across 22,000 classes of objects and things organized by everyday English words. 在2009年,ImageNet项目诞生了—— 一个含有1500万张照片的数据库, 涵盖了22000种物品。 这些物品是根据日常英语单词进行分类组织的。
organized:adj.有组织的; v.组织; (organize的过去分词和过去式)
In both quantity and quality, this was an unprecedented scale . 无论是在质量上还是数量上, 这都是一个规模空前的数据库。
unprecedented:adj.空前的;无前例的; scale:n.规模;比例;鳞;刻度;天平;数值范围;v.衡量;攀登;剥落;生水垢;
As an example, in the case of cats, we have more than 62,000 cats of all kinds of looks and poses and across all species of domestic and wild cats. 举个例子,在“猫”这个对象中, 我们有超过62000只猫 长相各异,姿势五花八门, 而且涵盖了各种品种的家猫和野猫。
poses:姿势; species:n.[生物]物种;种类; domestic:n.佣人;家佣;家庭纠纷;家庭矛盾;adj.本国的;国内的;家用的;家庭的;
We were thrilled to have put together ImageNet, and we wanted the whole research world to benefit from it, so in the TED fashion, we opened up the entire data set to the worldwide research community for free. 我们对ImageNet收集到的图片感到异常兴奋, 而且我们希望整个研究界能从中受益, 所以以一种和TED一样的方式,我们公开了整个数据库, 免费提供给全世界的研究团体。
thrilled:adj.非常兴奋; v.使非常兴奋; (thrill的过去分词和过去式) put together:..放在一起;组合;装配; worldwide:adj.全世界的;adv.在世界各地; community:n.社区;[生态]群落;共同体;团体;
(Applause)
Now that we have the data to nourish our computer brain, we're ready to come back to the algorithms themselves. 那么现在,我们有了用来培育计算机大脑的数据库, 我们可以回到”算法“本身上来了。
nourish:vt.滋养;怀有;使健壮;
As it turned out, the wealth of information provided by ImageNet was a perfect match to a particular class of machine learning algorithms called convolutional neural network , 因为ImageNet的横空出世,它提供的信息财富完美地 适用于一些特定类别的机器学习算法, 称作“卷积神经网络”,
wealth:n.财富;大量;富有; provided:conj.假如; v.提供; (provide的过去分词和过去式) neural network:n.神经网络;
pioneered by Kunihiko Fukushima, Geoff Hinton, and Yann LeCun back in the 1970s and '80s. 在上世纪七八十年代开创。
Just like the brain consists of billions of highly connected neurons, a basic operating unit in a neural network is a neuron-like node . 就像大脑是由上十亿的紧密联结的神经元组成, 神经网络里最基础的运算单元 也是一个“神经元式”的节点。
consists:v.由…构成;由…组成(consist的三单形式); highly:adv.高度地;非常;非常赞许地; node:n.节点;结点;茎节;(根或枝上的)节;
It takes input from other nodes and sends output to others. 每个节点从其它节点处获取输入信息, 然后把自己的输出信息再交给另外的节点。
input:n.投入; v.把(数据等)输入计算机; nodes:n.茎节;(根或枝上的)瘤,节,结;节点;(node的复数) output:n.(人、机器、机构的)产量;输出;输出功率;输出量;v.输出;
Moreover , these hundreds of thousands or even millions of nodes are organized in hierarchical layers , also similar to the brain. 此外,这些成千上万、甚至上百万的节点 都被按等级分布于不同层次, 就像大脑一样。
Moreover:adv.而且;此外; hierarchical:adj.分层的;等级体系的; layers:n.层;表层;层次;阶层;v.把…分层堆放;(layer的第三人称单数和复数)
In a typical neural network we use to train our object recognition model, it has 24 million nodes, 140 million parameters , and 15 billion connections. 在一个我们用来训练“对象识别模型”的典型神经网络里, 有着2400万个节点, 1亿4千万个参数, 和150亿个联结。
typical:adj.典型的;特有的;象征性的; recognition:n.识别;认识;承认;认可; parameters:n.决定因素;规范;范围;(parameter的复数)
That's an enormous model. 这是一个庞大的模型。
Powered by the massive data from ImageNet and the modern CPUs and GPUs to train such a humongous model, the convolutional neural network blossomed in a way that no one expected. 借助ImageNet提供的巨大规模数据支持, 通过大量最先进的CPU和GPU,来训练这些堆积如山的模型, “卷积神经网络” 以难以想象的方式蓬勃发展起来。
humongous:adj.巨大无比的,极大的; blossomed:v.开花;变得更加健康(或自信、成功);(blossom的过去分词和过去式)
It became the winning architecture to generate exciting new results in object recognition. 它成为了一个成功体系, 在对象识别领域,产生了激动人心的新成果。
architecture:n.建筑学;建筑风格;建筑式样;架构;
This is a computer telling us this picture contains a cat and where the cat is. 这张图,是计算机在告诉我们: 照片里有一只猫、 还有猫所在的位置。
Of course there are more things than cats, so here's a computer algorithm telling us the picture contains a boy and a teddy bear; a dog, a person, and a small kite in the background; or a picture of very busy things like a man, a skateboard , railings , a lampost, and so on. 当然不止有猫了, 所以这是计算机算法在告诉我们 照片里有一个男孩,和一个泰迪熊; 一只狗,一个人,和背景里的小风筝; 或者是一张拍摄于闹市的照片 比如人、滑板、栏杆、灯柱…等等。
teddy:泰迪玩具熊 skateboard:n.溜冰板;vi.用滑板滑行; railings:n.[建]栏杆(railing的复数);栅栏;围栏;
Sometimes, when the computer is not so confident about what it sees, we have taught it to be smart enough to give us a safe answer instead of committing too much, just like we would do, 有时候,如果计算机不是很确定它看到的是什么, 我们还教它用足够聪明的方式 给出一个“安全”的答案,而不是“言多必失” ——就像人类面对这类问题时一样。
confident:adj.自信的;确信的; committing:v.做出;犯罪或错等;自杀;承诺,保证;(commit的现在分词)
but other times our computer algorithm is remarkable at telling us what exactly the objects are, like the make, model, year of the cars. 但在其他时候,我们的计算机算法厉害到可以告诉我们 关于对象的更确切的信息, 比如汽车的品牌、型号、年份。
remarkable:adj.卓越的;非凡的;值得注意的;
We applied this algorithm to millions of Google Street View images across hundreds of American cities, and we have learned something really interesting: first, it confirmed our common wisdom that car prices correlate very well with household incomes. 我们在上百万张谷歌街景照片中应用了这一算法, 那些照片涵盖了上百个美国城市。 我们从中发现一些有趣的事: 首先,它证实了我们的一些常识: 汽车的价格, 与家庭收入呈现出明显的正相关。
applied:adj.应用的;实用的;v.应用;使用;申请,请求;(apply的过去分词和过去式) Google:谷歌;谷歌搜索引擎; wisdom:n.智慧;明智;才智;学问; correlate:vi.关联;vt.使有相互关系;互相有关系;n.相关物;相关联的人;adj.关联的;
But surprisingly , car prices also correlate well with crime rates in cities, or voting patterns by zip codes. 但令人惊奇的是,汽车价格与犯罪率也呈现出明显的正相关性, 以上结论是基于城市、或投票的 邮编区域进行分析的结果。
surprisingly:adv.令人惊讶地;出乎意料地
So wait a minute. Is that it? 那么等一下,这就是全部成果了吗？
Has the computer already matched or even surpassed human capabilities? 计算机是不是已经达到,或者甚至超过了人类的能力？
surpassed:v.超过,凌驾(surpass的过去分词形式);
Not so fast. ——还没有那么快。
So far, we have just taught the computer to see objects. 目前为止,我们还只是教会了计算机去看对象。
This is like a small child learning to utter a few nouns . 这就像是一个小宝宝学会说出几个名词。
utter:v.说;出声;讲;adj.完全的;十足的;彻底的; nouns:名词(noun的复数)
It's an incredible accomplishment , but it's only the first step. 这是一项难以置信的成就, 但这还只是第一步。
incredible:adj.难以置信的,惊人的; accomplishment:n.成就;完成;技艺,技能;
Soon, another developmental milestone will be hit, and children begin to communicate in sentences. 很快,我们就会到达发展历程的另一个里程碑: 这个小孩会开始用“句子”进行交流。
milestone:n.里程碑,划时代的事件;
So instead of saying this is a cat in the picture, you already heard the little girl telling us this is a cat lying on a bed. 所以不止是说这张图里有只“猫”, 你在开头已经听到小妹妹告诉我们“这只猫是坐在床上的”。
So to teach a computer to see a picture and generate sentences, the marriage between big data and machine learning algorithm has to take another step. 为了教计算机看懂图片并生成句子, “大数据”和“机器学习算法”的结合需要更进一步。
Now, the computer has to learn from both pictures as well as natural language sentences generated by humans. 现在,计算机需要从图片 和人类创造的自然语言句子中同时进行学习。
as well as:也;和…一样;不但…而且; natural language:n.自然语言(自然发展而成,并非人造);
Just like the brain integrates vision and language, we developed a model that connects parts of visual things like visual snippets with words and phrases in sentences. 就像我们的大脑,把视觉现象和语言融合在一起, 我们开发了一个模型, 可以把一部分视觉信息, 像视觉片段,与语句中的文字、短语联系起来。
integrates:v.(使)结合;(使)一体化;(使)合并(integrate的第三人称单数形式); snippets:n.片段(snippet的复数形式);小片;
About four months ago, we finally tied all this together and produced one of the first computer vision models that is capable of generating a human-like sentence when it sees a picture for the first time. 大约4个月前, 我们最终把所有技术结合在了一起, 创造了第一个“计算机视觉模型”, 它在看到图片的第一时间, 就有能力生成类似人类语言的句子。
finally:adv.终于;最终;(用于列举)最后;彻底地; capable:adj.能干的,能胜任的;有才华的; generating:v.产生;引起;(generate的现在分词)
Now, I'm ready to show you what the computer says when it sees the picture that the little girl saw at the beginning of this talk. 现在,我准备给你们看看计算机看到图片时 会说些什么 ——还是那些在演讲开头给小女孩看的图片。
at the beginning of:在…的开始;
(Video) Computer: A man is standing next to an elephant. （视频）计算机:“一个男人站在一头大象旁边。”
A large airplane sitting on top of an airport runway . “一架大飞机停在机场跑道一端。”
airport:n.机场;航空港; runway:n.跑道;河床;滑道;
FFL: Of course, we're still working hard to improve our algorithms, and it still has a lot to learn. 李飞飞:当然,我们还在努力改善我们的算法, 它还有很多要学的东西。
improve:v.改进;改善;
(Applause) （掌声）
And the computer still makes mistakes. 计算机还是会犯很多错误的。
(Video) Computer: A cat lying on a bed in a blanket. （视频）计算机:“一只猫躺在床上的毯子上。”
FFL: So of course, when it sees too many cats, it thinks everything might look like a cat. 李飞飞:所以…当然——如果它看过太多种的猫, 它就会觉得什么东西都长得像猫……
(Video) Computer: A young boy is holding a baseball bat. （视频）计算机:“一个小男孩拿着一根棒球棍。”
(Laughter) （笑声）
FFL: Or, if it hasn't seen a toothbrush , it confuses it with a baseball bat. 李飞飞:或者…如果它从没见过牙刷,它就分不清牙刷和棒球棍的区别。
toothbrush:n.牙刷; confuses:v.使糊涂;使迷惑;(将…)混淆;使更难于理解;(confuse的第三人称单数)
(Video) Computer: A man riding a horse down a street next to a building. （视频）计算机:“建筑旁的街道上有一个男人骑马经过。”
(Laughter) （笑声）
FFL: We haven't taught Art 101 to the computers. 李飞飞:我们还没教它Art 101（美国大学艺术基础课）。
(Video) Computer: A zebra standing in a field of grass. （视频）计算机:“一只斑马站在一片草原上。”
FFL: And it hasn't learned to appreciate the stunning beauty of nature like you and I do. 李飞飞:它还没学会像你我一样欣赏大自然里的绝美景色。
appreciate:v.欣赏;感激;感谢;理解; stunning:adj.惊人的; v.使昏迷; (stun的现在分词)
So it has been a long journey. 所以,这是一条漫长的道路。
To get from age zero to three was hard. 将一个孩子从出生培养到3岁是很辛苦的。
The real challenge is to go from three to 13 and far beyond. 而真正的挑战是从3岁到13岁的过程中,而且远远不止于此。
Let me remind you with this picture of the boy and the cake again. 让我再给你们看看这张关于小男孩和蛋糕的图。
remind:v.提醒;使想起;
So far, we have taught the computer to see objects or even tell us a simple story when seeing a picture. 目前为止,我们已经教会计算机“看”对象, 或者甚至基于图片,告诉我们一个简单的故事。
(Video) Computer: A person sitting at a table with a cake. （视频）计算机:”一个人坐在放蛋糕的桌子旁。“
FFL: But there's so much more to this picture than just a person and a cake. 李飞飞:但图片里还有更多信息—— 远不止一个人和一个蛋糕。
What the computer doesn't see is that this is a special Italian cake that's only served during Easter time. 计算机无法理解的是:这是一个特殊的意大利蛋糕, 它只在复活节限时供应。
The boy is wearing his favorite t-shirt given to him as a gift by his father after a trip to Sydney, and you and I can all tell how happy he is and what's exactly on his mind at that moment. 而这个男孩穿着的是他最喜欢的T恤衫, 那是他父亲去悉尼旅行时带给他的礼物。 另外,你和我都能清楚地看出,这个小孩有多高兴, 以及这一刻在想什么。
This is my son Leo. 这是我的儿子Leo。
On my quest for visual intelligence, 在我探索视觉智能的道路上,
quest:n.追求;寻找;vi.追求;寻找;vt.探索;
I think of Leo constantly and the future world he will live in. 我不断地想到Leo和他 未来将要生活的那个世界。
When machines can see, doctors and nurses will have extra pairs of tireless eyes to help them to diagnose and take care of patients . 当机器可以“看到”的时候, 医生和护士会获得一双额外的、不知疲倦的眼睛, 帮他们诊断病情、照顾病人。
extra:adj.额外的:n.额外的事物:adv.额外:另外: tireless:adj.不知疲倦的;不疲劳的; diagnose:vt.诊断;断定;vi.诊断;判断; patients:n.接受治疗者,病人;(patient的复数)
Cars will run smarter and safer on the road. 汽车可以在道路上行驶得更智能、更安全。
Robots, not just humans, will help us to brave the disaster zones to save the trapped and wounded . 机器人,而不只是人类, 会帮我们救助灾区被困和受伤的人员。
disaster:n.灾难,灾祸;不幸; trapped:adj.受困的;受限制的;v.使落入险境;使陷入困境;(trap的过去分词和过去式) wounded:adj.受伤的; n.伤员; v.使受伤; (wound的过去分词和过去式)
We will discover new species, better materials, and explore unseen frontiers with the help of the machines. 我们会发现新的物种、更好的材料, 还可以在机器的帮助下探索从未见到过的前沿地带。
explore:v.探索:探测:探险: unseen:adj.看不见的,未看见的;未经预习的;n.(事前未看过原文的)即席翻译;
Little by little , we're giving sight to the machines. 一点一点地,我们正在赋予机器以视力。
Little by little:渐渐;逐渐地;
First, we teach them to see. 首先,我们教它们去“看”。
Then, they help us to see better. 然后,它们反过来也帮助我们,让我们看得更清楚。
For the first time, human eyes won't be the only ones pondering and exploring our world. 这是第一次,人类的眼睛不再独自地思考 和探索我们的世界。
pondering:v.沉思;考虑;琢磨;(ponder的现在分词) exploring:v.探索:考察:探查;(explore的现在分词)
We will not only use the machines for their intelligence, we will also collaborate with them in ways that we cannot even imagine. 我们将不止是“使用”机器的智力, 我们还要以一种从未想象过的方式,与它们“合作”。
collaborate:vi.合作;勾结,通敌;
This is my quest: to give computers visual intelligence and to create a better future for Leo and for the world. 我所追求的是: 赋予计算机视觉智能, 并为Leo和这个世界,创造出更美好的未来。
Thank you. 谢谢。
(Applause)