I recreated famous album covers with DALL-E

DALL-E is still highly probabilistic in its judgement. For instance, in this article, it keeps putting &quot;fire&quot; in the the background on something that is likely to be on fire, rather than lighting up the person.<p>I have a similar experience. In my own experiment, I can&#x27;t get DALL-E to turn off the street lamp at a bus stop in the darkness. I&#x27;ve tried &quot;no light&quot;, &quot;broken street lamp&quot;, etc.; no use. Any mention of &quot;street lamp&quot;, the scene will include a working street lamp.<p>It&#x27;s just more probable that a scene with a lamp in the darkness must have that lamp providing light, and this is something that DALL-E will not break out of.

Worth noting that DALL-E automatically “rejects attempts to create the likeness of any public figures, including celebrities&quot;. So, you wouldn&#x27;t be able to get an image that included the 4 Liverpudlians. It does allow you to create fake faces. Might be fun to try and recreate Miles Davis Tutu, Aladdin Sane, Piano Man.

Man, after seeing Stable Diffusion&#x27;s output, DALL-E&#x27;s looks just janky. Like watching a propeller plane after seeing a jet.<p>Crazy how fast the tech is moving.

After getting access to the beta, combined with all these HN posts -- I&#x27;ve determined DallE2 is neat but no where as great as the initial samples made me believe.

An upvote for whoever can give me a prompt to generate an image of someone who&#x27;s been massaged so much their body has been flattened, as if they were made of dough or jelly or something.<p>I spent ages on this earlier getting nowhere. I&#x27;m starting to think DALL-E is better if you don&#x27;t really know what you want and you&#x27;re just fishing for ideas.

Have anyone given a prompt to Dall-e of designing a company website and included “make it pop!”?<p>Maybe the AI will finally get what designers always complained about annoying clients.

I love seeing people experiment with this technology. You can feel we’re on the cusp of something great - whatever it is, we’re just not quite there yet.

&gt; <i>The question is, when the music blows up and the artwork becomes a signature, like the Rolling Stones&#x27; Tongue &amp; Lips, who will own the copyright?</i><p>That’s what trademarks are for.

How do you know that the album covers are not part of the corpus of images that DALL-E was trained on in the first place?

Interestingly related, I just used AI image generation to create my EP cover.. first I tried running luciddrains dall e 2 PyTorch implementation using the prompt “death by Tetris EP album cover 2022” unfortunately I am using a Mac Pro so the gpu was not able to work. Then I tried imagen PyTorch implementation and used same keyword. This time it was working with the CPU unfortunately 2 days in we had a power outage so I had something but nothing complete. So I fed the generated image into the google dream generator and got my album cover!<p><a href="https:&#x2F;&#x2F;willsimpson.hearnow.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;willsimpson.hearnow.com&#x2F;</a>

There are a lot of articles focusing on how close does DALL-E match some prompt, but I wonder if this is a suboptimal way to explore the medium.<p>What if you can get a lot more out of it by embracing the unexpected responses. Can it be a tool for exploring lateral thinking? You provide a prompt computer responds with images that are a prompt to human. A baby swiming next to a dolar bill outputs a distorted person face inside a dolar bill with some baby features, could be the start to a rabbit hole of prompts and images where you&#x27;ll end up with something completly different than your initial expectations.

It&#x27;s interesting that the prompts that would do badly in a Google image search also seem to be the ones that make poor prompts. Basically, it seems that rather than describing a scene, you have to try to give an analogy for some image(s) that it might have in its training set - which is why, I believe, &quot;banana in the style of Andy Warhol&quot; produces a much higher quality result than &quot;Outline of prism on a black background in the middle of scene splits a beam of light coming from the left side into rainbow on the right side&quot;.

Although AI artists will destroy a lot of jobs, it will also create demand for new jobs for people who specialize in “paint overs” – taking a high concept output created by AI artists and touching it up to perfection.<p>Or perhaps even beyond just a paint over, and into the realm of recreating an entire AI artwork but with a human touch to get details just right.<p>Looking forward to it.

DALL E works really well if you are specific enough. When you don&#x27;t get the intended result, it helps to identify the element which wasn&#x27;t generated properly and then improve your description of the same.<p>&quot;Two men, one of whom is on fire, shaking hands on an industrial lot.&quot; can be rewritten as, &quot;Two men, shaking hands, standing on an industrial lot. Person on the right is on fire. Camera is 30 metres away.&quot;<p>You can go into more specifics of the framing and the angle from which you want the picture to be take. By default, DALL E will give you the most realistic generations to your prompts unless you mention &quot;digital art&quot; or a particular art style. I have gotten the best results when generating art instead of photos.

I haven&#x27;t gotten to try it for myself, but I&#x27;ve read a few of these blogs that take you through generating examples or even look-alikes to older art pieces.<p>It surprisingly reminds me a lot of when I traveled to Japan without knowing really any Japanese. I needed to communicate not only with friends who don&#x27;t know much English either, but also other people (like restaurant wait staff, train station staff, etc.).<p>I used Google Translate often, but many times I or my friend(s) (or the other people) would need to re-write our statements a few times until the translation result clicked well enough in each other&#x27;s languages to be understandable.

Is the issue with faces a deliberate choice by the devs?

I&#x27;ve been recreating the 50 worst heavy metal album art using AI as well, currently at 30. Recently I&#x27;ve found Stable Diffusion plus DALL-E inpainting to be a good combination.<p><a href="https:&#x2F;&#x2F;twitter.com&#x2F;P_Galbraith&#x2F;status&#x2F;1560469019605344256" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;P_Galbraith&#x2F;status&#x2F;1560469019605344256</a>

Very cool. But it just goes to show the impact of human creativity. The conceptual aspect.

No Smell the Glove cover, this is a black day for rock and roll!

If you gave those same instructions to humans I’m sure the output would be just as varied. I’d be interested to see a comparison between dall-e and humans.

I&#x27;d love to see what it had come up with if simply prompted for &quot;Album cover for Nevermind by Nirvana&quot;

I wonder, how much energy is being burned for these kinds of experiments.

It&#x27;s going to leave all those artists without a job, you just wait!!

Went to use my invite, and OpenAI demands your PHONE NUMBER.<p>No excuse for it. Screw that.

Look, it&#x27;s trained on these images.<p>It&#x27;s really great and cool and all - but it&#x27;s retrieving things that it was trained on.<p>Show me something original it did.

Is i do smth with DALL-E auto top hacker news post i saw like 20 post like that in past 2 weeks.

DALL-E still seems very useless. Reminds me of the hype of Cardano.

I wonder how long the novelty of DALL-E will persist. HN seems to upvote anything titled &quot;I did X with DALL-E&quot;. This is a fun post, but it&#x27;s not that interesting or surprising. Still worth a look don&#x27;t get me wrong, but personally didn&#x27;t learn anything new from it. (eg recreating the famous pink Floyd cover with &quot;Outline of prism on a black background in the middle of scene splits a beam of light coming from the left side into rainbow on the right side&quot; unsurprisingly worked somewhat well).