but we know the internet is biased toward the english language in western concepts with whole culture is not represented at all in one open source dataset the word asian is represented first and foremost by an avalanche of porn a region that captures the essence of banananus a region that represents the textures and colors of photos from the 1960s and area for snow and an area for globes and snow globes somewhere in between any point in this space can be thought of as the recipe for a possible image and the text prompt is what navigates us to that location but then there's one more step translating a point in that mathematical space into an actual image in pixel space involves, a generative process called diffusion it starts with just noise and then over a series of iterations arranges pixels into a complex position that makes sense to humans because of some randomness in the process, it will never return exactly the same image for the same prompt and if you enter the prompt into a different model designed by different people and trained on different data, you'll get a different result because you're in a different latent space okay, no way that is so cool what the heck the like brush strokes the color palette that's fascinating i wish i could like i mean he's dead but like got to be like look what i oh that's pretty cool? probably, the only dolly that i could afford any ice the ability of deep learning to extract patterns from data means that you can copy an artist style without copying their images just by putting their name in the prompt, james gurney is an american illustrator who quickly became a popular reference for users of text to image models i asked him what kind of norms? he would like to see as prompting becomes widespread i think it's only fair to people looking at this work that they should know what the prompt was and also what software was used also i think the artist should be allowed to opt in or opt out of having their work that they worked so hard on by hand be used as a dataset for creating this other artwork james gurney i think he was a great example of being someone who was open to it started talking with their artists, but i also heard of all the artists who got actually extremely upset the copyright questions regarding the images that go into training the models and the images that come out of them are completely unresolved and those aren't the only questions that this technology will provoke the latent space of these models contain some dark corners that get scarier as outputs become photorealistic it also holds an untold number of associations that we wouldn't teach our children, but that it learned from the internet if you ask for an image of a ceo, it's like an old like guy if you ask for images of nurses, they're all like women we don't know exactly what's in the data sets used by open, ai or mid journey but we know the internet is biased where the english language in western concepts with whole culture is not represented at all in one open source dataset the word asian is represented first and foremost by an avalanche of porn yeah, it really is just sort of a infinitely complex mirror you know held up to our society and what we deem worthy enough to you know put on the internet in the first place and how we think about what we do put up but what makes this technology so unique is that it enables any of us to direct the machine to imagine what we wanted to see a party hat guy space invader, a caterpillar and a ramen bowl prompting removes the obstacles between our ideas and images in eventually videos animations and whole virtual worlds we are on a voyage here that is it's a bigger deal than just like one decade or immediate technical consequences it's a change in the way, humans imagine communicate work with their own culture and that will have long range good and bad consequences that we are just by definition not going to be capable of completely anticipating over the course of researching this video, i spoke to a bunch of creative people who have played with these tools and i asked them what they think this all means for people who make a living making images, the human artists and illustrators and designers and stock photographers out there and they had a lot of interesting things to say so i've compiled them into a bonus video, please check it out and add your own thoughts in the comments thank you for watching,
粉丝2568获赞1.1万

but we know the internet is biased where the english language in western concepts with whole culture is not represented at all in one open source dataset the word asian is represented first and foremost by an avalanche of porn mid journey currently has a waitlist for subscriptions but we got a chance to try it out okay, man go ahead and take a look oh wow that is so cool! it has some work to do i feel like it can be it's not dancing and it could be better the craft of communicating with these deep learning models has been dubbed prompt engineering what i love about prompting for me? it's kind of really it has something like magic where you have to know the right words for the spell in real life that you can refine the way? you talk to the machine it can become like kind of a dialogue you can say like octane render blender 3d made with unreal engine certain types of film lenses and cameras 1950s, 1960s diets diets are really good lino cut or wood cut coming up with funny pairings like fabrice egg mcmuffins a monochromatic infographic poster about typography depicting chinese characters some of the most striking images can come from prompting the model to synthesize a long list of concepts it's kind of like a it's having a very strange collaborator to bounce ideas off of and get unpredictable ideas back i love that my prompt was chasing seafoam dreams which is a lyric from the ted leo and the farmers this song by a musicalology can i use it as the album cover for my first album all right for an image generator to be able to respond to so many different prompts it needs a massive diverse training dataset like hundreds of millions of images scraped from the internet, along with their text descriptions, those captions come from things like the alt text that website owners upload with their images for accessibility and for search engines, so that's how engineers get these giant data sets, but then what do the models actually do with them? we might assume that when we give them a text prompt like a banana inside a snow globe from 1960 they search through the training data to find related images and then copy over some of those pixels, but that's not what's happening the new generated image doesn't come from the training data it comes from the latents space of the deep learning model that'll make sense in a minute first, let's look at how the model learns if i gave you these images and told you to match them to these captions, you'd have no problem but what about now? this is what images? look like to a machine just pixel values for red grain and blue you just have to make a guess and that's what the computer does to at first, but then you could go through thousands of rounds of this and never figure out how to get better at it whereas a computer can eventually figure out a method that works that's what deep learning does in order to understand that this arrangement of pixels is a banana and this arrangement of pixels is a balloon it looks for metrics that help separate these images in mathematical space, so how about color if we measure the amount of yellow in the image, that would put the banana over here and the balloon over here in this one dimensional space, but then what if we run into this now, our yellowness metric isn't very good at separating bananas from balloons we need a different variable so let's add an axis for roundness now we've got a two dimensional space with the round balloons up here and the banana down here but if we look at more data, we may come across a banana that's pretty round in a balloon that isn't so maybe there's some way to measure shininess balloons usually have a shiny spot now we have a three dimensional space and ideally when we get a new image, we can measure those three variables and see whether it falls in the banana region or the balloon region of the space for what if we want our model to recognize not just bananas and balloons, but all these other things, yellowness, roundness and shininess don't capture what's distinct about these objects we need better variables and we need a lot more of them learning algorithms do as they go through all the training data they find variables that help improve their performance on the task and in the process they build out a mathematical space with way more than three dimensions we are incapable of picturing multi dimensional space but mid journey's model offered this and i like it so we'll say this represents the latent space of the model and it has more than 500 dimensions those 500 axes represent variables that humans wouldn't even recognize or have names for, but the result is that the space has meaningful clusters。

but we know the internet is biased where the english language in western concepts with whole culture is not represented at all in one open source dataset the word asian is represented first and foremost by an avalanche of porn seven years ago back in 2015, one major development in ai research was automated image captioning machine learning algorithms could already label, objects and images, and now they learn to put those labels into natural language descriptions and it made one group of researchers curious what if you flipped that process around, we could do image to text why not try doing text image as well and see how it works it was a more difficult task they didn't want to retrieve existing images the way google search does they wanted to generate entirely novel scenes that didn't happen in the real world so they asked their computer model for something that it would have never seen before like all the school buses you've seen are yellow but if you write the red or green school bus, would it actually try to generate something green and it did that it was a 32 by 32 tiny image and then all you could see is like a blob of something on top of something they tried some other prompts like a herd of elephants flying in the blue skies, a vintage photo of a cat, a toilet seat sits open in the grass field and a bowl of bananas is on the table maybe not something to hang on your wall but the 2016 paper from those researchers showed the potential for what might become possible in the future and the future has arrived it is almost impossible to oversee how far the technology has come in just one year leaps and bounds yeah, oh, it's been quite dramatic i don't know anyone who hasn't immediately been like what is this? what is happening here? so could i say like waves crushing watching waves crushing okay party hat guy seafoam dream a coral reef cubism caterpillar a dancing tackle my prompt is salvador dahli painting the skyline of new york city you may be thinking wait ai generated images aren't new you might have heard about this generated portrait going for over four hundred thousand dollars at auction back in 2018 or this installation of morphing portraits, which thought to be sold the following year it was created by my mario klingeman who explained to me that that type of ai art required him to collect a specific dataset of images and train his own model to mimic that data let's say oh, i want to create landscapes so i collect a lot of landscape images i want to create portraits i trained it on portraits but then the portrait model would not really be able to create landscapes and same with those hyper realistic fake faces that have been plaguing linkedin in facebook those come from a model that only knows how to make faces generating a scene from any combination of words requires a different newer bigger approach yeah now we kind of have these huge models which are so huge that somebody like me actually cannot train them anymore on their own computer, but once they are there they are really kind of they contain everything i mean to a certain extent means is that we can now create images without having to actually execute them with paint or cameras or pen tools or code the input is just a simple line of text and i'll get to how this technology works later in the video, but to understand how we got here we have to real wind to january 2021 that's when a major ai company called open, ai announced dolly, which they named after these guys, they said that it could create images from text captions for a wide range of concepts and they recently announced dolly too which promises more realistic results and seamless editing, but they haven't released either version to the public so over the past year a community of independent open source developers built text to image generators out of other pre trained models that they did have access to and you can play with those online for free some of those developers are now working for a company called mid journey, which created a discord community with bots that turn your text into images in less than a minute having basically, no barrier to entry do this has made it like a whole new ball game i've been up until that tour free in the morning just you know really china changed things and piece things together, and it's on about seven thousand images it's just ridiculous mid journey currently has a weight list for subscriptions, but we got a chance to try it out,

在商界哈,一直以来都流传着这么一个思想,就是你在淘金巢的时候,你费了半天劲去挖着那个金子,其实是个非常小概率的事件,但是你有一个稳赚不赔的操作,就咱不去挖那个金子,咱去卖铲子。 金子在哪?虽然咱不知道,但咱知道的是,挖金子的人肯定会大步大步的涌过来,那我就把铲子卖给他。那二零二三年这个所谓的淘金潮是什么呢?就是人工智能吧,虽然人工智能战场现在还是扑朔迷离,短短两三个月就冒出来几十个大模型,但有一个卖铲子的公司,那可是笑的最开心的,市值甚至超过了特斯拉,那就是英伟达, 他卖的这个所谓的铲子啊,就是显卡,而所有的大模型都得靠显卡来训练,而英伟达呢,就几乎是唯一那个卖铲子的人,你看这钱赚多爽。了解一点经济学的人可能都知道啊,你说你要在淘金朝的时候卖铲子,可能会赚大钱吗?第一个去的可能会赚点钱,可慢慢的大家伙都来卖了,铲子又不难做,你的边际利润就会迅 迅速被吞噬。所以在大部分的行业里,你去卖这个所谓的铲子,就是那个上游的生产工具,因为他主入门槛太低,行业竞争就会很激烈,所以利润率根本不高。 可是烟味达呢,他就能靠做这个所谓的铲子做到垄断,做到万一失职,包括台阶店。有点类似啊,他比铲子还基础,就做锤子,就是这个做铲子的工具,这也能垄断?你想没想过是为什么? 哼!大家都听说过摩尔定律吧,就是集成电路上可容纳的晶体管的数目,每十八个月就能翻一倍,或者你可以很笼统的理解成芯片每十八个月就能快一倍。黄荣勋就发现哈,这批有独立显卡, 他的推进速度好像比这个摩尔定律预测的还要快,大概每两年性能就能翻超过三倍。而这个规律甚至都有了他自己的专有名词叫皇室定律,慌死了。不管哪个定律哈,这个就是芯片行业他非常显著的一个特点,他确实初期的成本投入非常高,需要大量的人才和设备, 更麻烦的是,他的更新迭代速度实在是太快,这是一个不停在奔跑的行业,行业更新速度太快。从商业角度讲,哈,对这个参与者来讲,有很棘手的问题,就是他很难构建自己的护城河。 你像传统公司,我花个大价钱,建个工厂,建个铁路,那就非常强的规模优势,成本优势,那互联网公司护城河就更强了,一旦形成网络效应,那微信、抖音、 facebook 对吧?用户就是你极强的护城河。 这芯片行业呢?那不是有摩尔定律,皇室定律,不管你今天推出来一个多炸的芯片,两年之后他注定已经落后,就你也不知道明年从哪来个毛头小子,带着什么新技术就把你给干翻了。 这个行业里边,你很难靠一个产品或者一个技术本身形成一个超过两年的护城河。当然确实你可以慢慢积累一堆专利来进行自我保护。可是事实证明,就大部分情况下啊,人家还是有办法能绕过你这堆专利。这个保护能力他并不强,你只能 不得不持续的奔跑,跑的比所有人都快。这个奔跑能力本身才是你的护城河。放到芯片行业呢,其实就是你的研发能力,就跟着研发这一整套的人才配套设施,组织架构,这个才是芯片公司的护城河。 为什么今天研发贵啊?一是他初期成本就很高,需要大量的人才,更关键的是你还得持续不停的快速迭代,快速的奔跑,这给一般人谁受得了? 你就看那些互联网巨头,他们已经非常注重科技创新了吧,什么亚马逊、谷歌、微软,研发投入其实也就营收了百分之十到十五,可是英伟达呢?他就要持续不断的投入百分之二十五左右的营收来继续研发。你以为那黄氏定律是什么自然规律吗?那还不是黄日勋自己卷自己的结果吗? 我们要是讲故事,可以说英伟达零六年之后一直垄断显卡市场这么一句话就完了,实际上他们好几次都是完全颠覆了自己之前的框架和技术,就逼着自己差不多每六个月就得推出新 新一代的芯片。这种激烈奔跑速度,使得这个行业里必须得是跑的最快,人才储备最充沛,最财大气粗,那么两三个公司才能赚到钱,其他人肯定得陪跑。这也是为什么我们经常看到芯片领域一顿竞争之后,经常是以收购收场。这里边不光是芯片研发难度的问题,更主要是他不经济,就连人才资本都非常充足,那些互联网巨头 逼到一定份上啊,也不敢轻易踏足芯片领域,哎,可是现在就不一样了, ai 领域可是互联网巨头接下来可能的最大战场之一,而这个战场里边显卡那就太重要了,毕竟显卡决定了计算速度,哪个巨头也不希望之后在 ai 都普及的时代被英伟打卡脖子是吧? 所以说虽然他们现在还是英伟达最大的客户,但他们也都在紧锣密鼓的自己这边开始钻研搞研发灯芯片。谷歌已经开发出来专门针对人工智能训练的芯片,叫 tpu, 据他们自己说哈比英伟达的显卡计算效率还要高。迈特其实早在二零一七年就开始用两 两万多台英伟达的显卡来进行 ai 训练,不过现在他们也要开始凹印自研芯片,简称叫 mtia。 其实二零二零年开始啊,英伟达就一直积极的想找孙正义买一下他旗下这个芯片公司二,但是像高通、微软、谷歌这些剧,我就纷纷叫出来强力反对,结果这位说我就没谈成你可见这些巨头其实是非常忌惮英伟达的。 所以你别看现在英伟达是风口上飞的最高的那个,你就觉得他走运,人家可是自己提前十多年花了那么多钱,费了那么大劲,自己把翅膀给做好,就站在风口那等着风来。

最近啊,首批因 ai 失业的人出现了,一些游戏公司正在大量裁员原画师,有的公司已裁掉一半人以上。原画师的工作就是绘画游戏的人物、环境和道具的。 原画师之所以被裁,是因为游戏公司引入 ai 后,工作效率至少能提升百分之五十,而且画稿质量也跟原画师不相上下。那这个消息给了我们哪些启示呢?第一点启示是, ai 时代正在重新定义简单和复杂。 过去我们认为保姆、月嫂这类工作简单,因为门槛低。而程序员、原画师这类工作复杂,因为门槛高。但这套逻辑正在过时。未来评价一份工作简单还是复杂,主要是看他的规则是否明确。你比如说原画师绘画规则就很明确, 就是根据策划的文案,绘制一个个特定的角色,如果把一个人物的性别、种族、职业、坐骑等,把这些描述都告诉 aiai, 就能画出来。再比如,我们知道 ai 下围棋很厉害,其实原因也在于围棋的规则很明确, 大家知道规则后,不断训练,记忆就会突飞猛进。相反,保姆、月嫂这类工作规则就不明确,做得好不好,很大程度是看人的主观感受。你比如说,有些工作涉及到交流,当对方心情低落时,你能否理解他的心情,并说出宽慰的话, 这看似不难,但很讲究恰当。什么是恰当呢?就是用你的同理心去判断此时该做什么,该说什么,或者该不该说话。这些都没有一定之规,全凭人的经验 和感觉。但对 ai 来说,就是规则不明确,他会感到力不从心。第二点启示是,未来没有哪份工作会是铁饭碗。首先, ai 取代部分人力将发生在各行各业。未来的工作形态大多会是人和 ai 的写作, 比如程序员、编程环节,规则很明确,未来 ai 会比人更擅长。站在理解项目经理需求的这件事上,没有明确的规则,人比 ai 更擅长,所以两者相互协作。再比如律师,在文件搜索、分类等环节上, ai 比人更擅长。 但在理解复杂的法律概念上,人比 a 更擅长,所以两者同样也是互相协作。但协作就意味着部分人力会被取代,一个原本三十人的团队很有可能就会缩减为十五人。未来这样的现象会发生在各行各业,只是 a 取代的程度不同,造成的失业比例不同。其次,即使有少数行业 ai 难以介入,但随着其他行业失业人增多了,他们再来找工作时,你所在的行业也会变得更卷了。假如你是一名理发师, 哎,不见得能直接抢你的饭碗,但他很有可能会抢一名会计的饭碗。然后那名会计可能转行做理发,又来和你争饭碗了。 所以,未来我们要么跟 ai 竞争,要么跟人竞争,没有谁敢说自己的职业是个铁饭碗。那普通人该怎样做到避免被淘汰呢?这里有两点建议。第一,是啊,努力提高自己的情商。 当 ai 在智力层面逐步赶上人类,情商可以说是最后的堡垒。这里的情商是指理解他人的情感和诉求,做到恰如其分的回应。比如很多人 都说,哎,现在都能写小说写论文了,文字工作者将要被淘汰。这里说的文字工作者该不如说是资料收集员。要知道,文字是人类思想的载体,请问有价值的思想会被 ai 定义取代吗?更重要的是,文字工作者的真正功底在于懂得什么是恰当。 一个好的作者能时刻想象不同读者读到一句话的反应,他写的东西有深度和温度,这背后是一个人的经历、阅历和思考。这些内功会被 ai 轻易取代吗? 同样,不论你从事哪项职业,都需要在情商方面努力。很多朋友爱问自己职业是否安全,这其实更多取决于你自己。未来,也许普通的咨询师会被淘汰,但很会与人沟通,很受人信赖。咨询师未必被淘汰。未来,也许普通的博物馆讲解 会被淘汰,但风趣幽默、很受人喜欢的讲解不仅不会被淘汰,还可能成为王牌讲解员。另一个建议是养成问为什么的喜欢人。相比 ai, 最可贵之处就在于人会问为什么人会关心因果性,而 ai 只关注相关性。 举例来说吧,尽管 ai 看病误诊率比人更低,但当你拿到一份 ct 图,问 ai 是如何判断肿瘤的,他只能告诉你是根据大数据推测出来,说白了就是跟肿瘤患者的 ct 图很像, 这就是相关性。但他给不出具体的解释,只有医生可以。因为医生懂得肿瘤形成内在逻辑,他能给病人讲清楚。所以 ai 再强大,也只能辅助医生做判断。医生如此,各行各业亦如此。对待一份工作,不要只是机械 化的执行,而是要保持思考,多问几个为什么,当你对行业研究足够,你就能驾驭 ai, 让 ai 帮你赋能,而不是被 ai 所架空。喜欢就请点赞关注吧!


如何判断一张图片是不是 ai 生成的?有一个简单的办法,那就是看图片里的手部,因为 ai 总是做出这样的这样的或是这样的手。那么为什么 ai 在生成手部的时候失误率更高?或者说 ai 为什么做不好手呢? 首先,我们都知道, ai 的绘画技术主要依赖于大量的数据训练,但是对于手指的数据比较缺乏,手指的形态、纹理、运动等方面的数据很难收集,因此难以训练出准确的模型。其次,人手指的形态和细节非常复杂,包括指甲、指纹、 皮肤褶皱等,这些细节的渲染需要高精度的模型和算法才能实现。最后,除了技术层面的问题,手指的绘制还涉及到人类的艺术审美。手指的细节和形态需要与整体画面融合,符合人类审美标准,这种审美难以被算法模拟,所以目前 ai 没办法很好的完成对手部的创作。

ar 绘画为什么被抵制?对于绘画行业的从业者而言,他们可能会感到担忧,如果这样的应用越来越普及,将会大大降低他们的工作价值。 毕竟,一个掌握绘画技能的画家凝聚了长期的学习和辛勤工作,而 ar 绘画几乎可以秒杀他们的艺术成果。 这就好比一个烹饪大师十年磨一剑,却被一个厨房机器人轻松取代,让他们感受到了被智能挤压的威胁。还有一些艺术品赏析者表示, ai 绘画缺乏作品的灵魂和独特性,会削弱人类对艺术真正的认识和欣赏。 他们的艺术创作纯粹是由算法生成的,没有怎么考虑与人之间的交流,尤其是他们无法表达出人类内心的情感、思想和感受。这就 如同一篇由机器生成的新闻报道,只是简单的事实陈述,而缺乏了人类独有的敏锐性和判断力,让人们难以产生共鸣和深层次的理解。还有人质疑, ai 绘画可能会带来一些道德和法律上的问题, 比如应用中可能会引入一些侵犯他人利益的素材、版权问题等,而这些都会损害到整个产业链和社会的正常秩序。 同样,在进行人像绘画中, ai 技术还存在个人隐私泄露的问题,用户不想被别人利用他们的面部信息进行侵犯, 这就像一个未经批准的私人侦探一样,侵犯了公民的安全和自由,让人们感到恐惧和不安。明白了这些现象背后的深层原因,我们或许可以更好地评估 ai 绘画技术的优 缺点。虽然人们对他们怀有担忧和抵制情绪,但我们也不妨通过更加全面的视角和多样化的思考方式来评估 ai 绘画的和颜悦色之处,进而为创造一个更加多元化和包容的艺术领域而努力。看没看懂?都欢迎你在评论区留言哦!


今天呢给大家介绍一个新模型, magic mix realistic v 六,名字很长是吧?它的中文名字呢,比较好听,叫做麦菊写实,你现在看到的这些图片呢,都是麦菊写实官方提供的视力图片, 通过这些图片我们不难发现,这个模型的生成能力还是很强的,除了传统的生成美女之外的,他也能够生成很多的其他的景物啊,劲舞啊,甚至动画人物,他还能够生成帅哥啊,动物啊,以及其他的一些斜视类的物体。 你可以通过下面的链接访问他在 cvtai 上的主页,那如果你要是不能科学上网的话,也可以通过下面的链接。那关于这个网站呢,我着重介绍一下,他的名字呢叫做利不利本 ai, 我们可以认为它是国内版的 secret ai, 里边同样包含大量的 ai 模型,而且 secret ai 上的资源呢,会议不定期的同步 录到这个网站上,所以说如果你没有办法去上 cbai 的话呢,这个网站呢也是一个不错的选择。那如何才能使用麦菊显示生成质量比较好的图片呢? 那作者呢,也提到了几点,那最重要的一点就是作者一直在强调一定不要使用人脸修复这个功能。其实不光是这个模型,我建议大家今后呢就尽量不要用这个功能了,因为那他的确挺鸡肋的。 在以往我们生成写实类的图片,我们的采样器呢,一般都是 dmp 系列的,但是在这里作者建议使用的却是欧拉系列的,如果你生成的是写实类的人物,那建议你用欧拉,那如果你生成的是非人物就是像警务啊,警务就用欧拉杠 a, 这里呢还有推荐的正向提示词和反向提示词,正向的提示词呢非常的常见,而负向的提示词呢则是两个 inviting, 这两个 inviting 呢也 比较常见,那一个是 n g d pronective, 另外一个呢就是白大汉的 v 四,大家可以通过下面这两个链接来下载这两个银白领。最后一点注意的就是作者建议将 skip clip 这个值呢设成二,那这个呢也是很多模型建议的一个技巧, 那掌握了这些技巧呢?下面呢我们就来实际生成一张图片来看一下。首先呢按照作者的建议编写正向的提示词,这个正向提示词呢应该说非常的简单,那反向提示词呢就更简单了,这呢我不用过多解释,采用算法,我们选择欧拉开启高清修复, 图片的宽高呢是从七六八乘五幺二点击生成,那如果你大致遵循这些规则的话,你生成的图片质量呢,应该都不会很差。那麦菊的第六版和第五版到底差在哪呢?那作者非常不负责任的说了一句,说第五版不 够写实,那剩下的你就得脑补了,那到底怎么叫不够写实呢?下面呢我们来对比一下,我们用相同的提示词和生成参数,用不同的模型来生成不同的图片,来对比一下第六版和第五版到底有什么区别。那我们先来看这组图片, 这组图片包含的元素比较多,是一个女骑士骑着马,后边还有连绵的群山,注意啊,别看脸,脸可以在后期修复的很漂亮,那我们主要看整体的呈现,那我们可以明显的看到,第六版的表现的更加写实一些,尤其是这匹马,第五版的明显有些比例失调, 而第六版的马呢,则会更俊美一些。第二组是美女的半身照,来对比的话呢,这两版的差距并不是很大。虽然说第五版不够写实,但是这种不够写实往往指的是景物或者静物,绝对不是人物。麦举,写实 主要就是用来生成人物的,如果他在人物上不够写实,其实呢,就有点差强人意了。第三组呢,是一组魔幻类的人物,他更像三地剑魔的作品,那通过这一组的对比呢,也能印证我们刚才的猜测,其实他的差别呢,不是很大。 第四组呢,实现面部的特写,这个呢,我们是能看到一些差距的,第六版的面部会表现的更加的精致和写实一些。注意啊,这两张脸呢,都没有经过高清修复。第五组呢,我想绘制一下中国的侠女,我想看一下中国人研发的模型,对于中国元素的表现到底怎么样。 通过对比我们不难发现,第五版在生成的这个图片里边,还是有一些西方的建筑和人物的影子,而第六版呢,明显就要好很多了, 真的让我想到了中国武侠世界里边那些白衣如雪,百趣如风的侠客。当然,这只是我对这个魔性的看法,他到底怎么样,你得自己用过才知道,关注我,让知识变得更有意思。

最近讨论最多的话题就是 ai 绘画,那么我们会不会被 ai 绘画所替代呢?首先, ai 绘画的出现对我们美术行业的冲击是非常非常大的, 也导致了一部分人面临着失业的危险,也加剧了我们这个行业的内疚程度。 他会替代一部分人,但是他替代不了我们普通人, 他会把那些把画画当成商品去服务别人的那部分人替代掉。因为 ai 绘画是能快速拿到一个结果的,而且这个结果几乎是完美的。但是我们 普通人画画的目的是为了释放情绪,舒减压力,表达自我, 这个过程是任何人任何产品都替代不了的。好吧,拜拜。

ai 绘画到底有什么魅力?这个人是谁?是不是感到有一丝陌生而又熟悉?我来告诉你,这个男人就是鼎鼎大名的莎士比亚。莎士比亚生前留下了不少肖像画,但是总觉得油画还是缺少一些真实的质感。 今天我们就利用 ai 的魔法复原出一张莎士比亚的真实照片。复原出的照片多了一丝真实的质感,更好的还原出了肤色。但是还不够,因为我们总可以看到莎士比亚穿着不合身的贵族服装,总让我有一些不好的联想。 我们再利用 ai 给莎士比亚换一套衣服。 最后,莎士比亚作为一个英国人,有着英国人从古至今的痛点, 那就是秃头。今天我们利用 ai 还莎士比亚一个满头秀发的梦,这就是 ai 绘画的魅力。

选择模板,输入文案,挑选形象,合成视频。大家好,我是玄机九九。之前我分享过一期使用 ai 数字人出镜代替口播的工具,喜欢的人挺多。最近我又发现了另一款数字人工具,它叫万兴播报,这是安卓、苹果和 windows, 最大的特点是多元化,非常适合做跨境电商、海外自媒体等出海内容。首先是角色多元化,比如说欧美风格的角色、亚洲角色,非洲角色既可以是帅哥,也可以是美女。 其次是配音和语言多元化,它拥有近百种不同的声线,汉语、英语、日语、葡萄牙语、法语、德语、韩语、泰语,世界上大部分流行的语言都可以在这使用。 ai 转换成语音 软件的使用非常简单,首先准备好一段文案,或者用智能脚板工具生成一段,然后首页选择一个模板,找到一个合适的数字人,把文案粘贴进去,可以按需要更换 背景和配音。最后点击导出,等待一会即可完成。 get ready for the ultimate beauty hall in my life stream tonight we're revealing the hottest makeup products for the six hundred and eighteen shopping festival。 除了内置的数字人和声音,他还支持私人订制克隆自己,满足更高级的需求。 总结一下,这是一个定位明确的爱工具,如果你有出海营销的需求,用半个实习生的话费定位 pro 版,可以免去找国外模特的时间和成本,连后期剪辑都省了下来,大大节省了运营成本。以上就是本期视频的全部内容,我们下期再见吧!