It's pretty easy to dismiss this stuff but I've found it has been very fun to play around with and certainly has allowed me to create imagery that far exceeds my skill.<p>Twitter seems to have an endless stream of examples with the aiart and generativeart hashtags. Also I really like Katherine Crowson's twitter because she builds Colab notebooks that n00bs like me can use to tinker:<p><a href="https://twitter.com/RiversHaveWings" rel="nofollow">https://twitter.com/RiversHaveWings</a><p><a href="https://github.com/crowsonkb" rel="nofollow">https://github.com/crowsonkb</a><p>Some examples I shared a while back using Katherine's VQGAN+CLIP notebook: <a href="https://imgur.com/a/1R5UZQb" rel="nofollow">https://imgur.com/a/1R5UZQb</a> Promp was 'mad max dieselpunk alien spaceship landed in a desert' and then I upscaled the images. These are cherry picked from a set of about 3x this size.<p>One thing I find pretty remarkable is how good it handles shadows and depth of field. These are all kind of mud tones but if you let it run away with color you'll get some really brilliant results.
I love using ru-DALLe to create artwork. I do several 3 hour runs a day. It's a "kind" of hobby in that it takes effort to collect the source materials and it takes effort to adjust the models to my liking.<p>Rather than calling these the final products, these are more like something I would take to an artist to produce some commissioned work in finer detail with their personal touch.<p><a href="https://imgur.com/a/zeCPnyA" rel="nofollow">https://imgur.com/a/zeCPnyA</a>
<a href="https://imgur.com/a/w2ifLVm" rel="nofollow">https://imgur.com/a/w2ifLVm</a>
<a href="https://imgur.com/a/gVlzlUG" rel="nofollow">https://imgur.com/a/gVlzlUG</a>
<a href="https://imgur.com/a/LOLq2ad" rel="nofollow">https://imgur.com/a/LOLq2ad</a>
<a href="https://imgur.com/a/QHyyrtY" rel="nofollow">https://imgur.com/a/QHyyrtY</a>
<a href="https://imgur.com/a/HFozuFe" rel="nofollow">https://imgur.com/a/HFozuFe</a>
<a href="https://imgur.com/a/eR6PTBR" rel="nofollow">https://imgur.com/a/eR6PTBR</a>
<a href="https://imgur.com/a/L7uKC8d" rel="nofollow">https://imgur.com/a/L7uKC8d</a>
<a href="https://imgur.com/a/kmOIQiW" rel="nofollow">https://imgur.com/a/kmOIQiW</a>
<a href="https://imgur.com/a/GWGt56i" rel="nofollow">https://imgur.com/a/GWGt56i</a><p>I have tons of these each day...though several of these examples use the same sources with different settings if you notice any monotony.
I love it. Unlike many machine learning examples this one does not highlight how good the model is, but highlights how poor the source material often is. Many of these are completely believable covers.
Typical AI generated art, only look like the real thing if you squint real hard or don't pay attention.<p>Very low value in my opinion, pretty much garbage in fact.
I love how the images include the wear and tear of old paperbacks giving these covers an authentic look... but I think the titles need some slight photoshopping to reflect those scratches :)
this artist has captured so many of the aesthetic qualities of these classic cover images here. One that is missing, I noticed, is the quality of sparking my imagination, or the evocation of a world. There's nothing in those images that is connecting with me on that kind of level.<p>Which makes me wonder if this is a problem with many/most of the AI images one sees. I think so?
This might have been useful to me in early eighties high school, I had a class in which I had to write book reports, once per week I think, and there was a sort of template as to how they had to be written with analysis of character motivations etc.<p>It's been a long time but I found something about the critical methodology used just extremely insulting to any piece of literature it was applied to, reductive and harmful to actually appreciating anything, so despite reading a minimum of 4 books a week I decided to write fake book reports. Generally crappy science fiction and fantasy was the genre.<p>I only remember one fake book I wrote a review of - The Hinterlands of Horlon - a crappy Tolkien rip off with a quest and everything, written by James R. Canton which was also a role playing character name I had for some occult spy game I was in at the time. Fake cover art would have been a nice topper to the fraud.
I find these kind of art kinda annoying to look at, because it takes like 3 or 4 secs until you realize that you are watching nonsense. The initial feeling fools me.
I believe that generative models are changing how content and art are created. Inspiration and judgement are becoming more important than skill:
In music, a good artist needed years of learning in theory and practice to become skilled at playing the instrument and creating harmonies. In the futures, the best artist are the ones that can be explore the input space of the AI. They can produce the most inspired prompts and judge best which results are good and how they can be changed.<p>The same can be said for painting (we are already seeing it now) and with writing.
The next step of course is for AI to do all the reading - as is actually already the case today. Bots that scan blogs/ websites and post artificial comments, together with a drop in reading among humans never seen before, with widespread functional illiteracy anong the newer generations.<p><a href="https://www.theguardian.com/education/2020/feb/29/children-reading-less-says-new-research" rel="nofollow">https://www.theguardian.com/education/2020/feb/29/children-r...</a>
My son put up a similar AI portal for his mother (who is a writer and artist). It takes a title and creates the art. For instance, I entered "Ford F150 by Frida Kahlo" and got a pastiche of red truck parts, faces and the letter 'F' followed by three or four quasi-symbols. All in a saturated style similar to Kahlo's paintings.<p>I'm guessing the faces because she did so much portraiture.
These covers aren't great, but if you could design a process that could produce great ones, then write books to fulfill the best ones, you wouldn't be that far off from the editorial process of a lot of pulp publishers that wrote books based on good covers.<p>AI generated covers and titles used as writing prompts would be very fun. They would have to be far better than these, though.
I notice the only cover with an author popular enough to have their name bigger than the title of the book is One Telling Ten Thousand by Isabella V. Rivera which does seem to go against the whole boys club feel of Science Fiction of that time. I think it would have been more likely a gender neutral name like Andre Norton's, maybe Kim Rivera or V. Rivera.
There's a part of me that actually wants to write a short story called "Green Glass is the Color of the Wind" now, since I think that's a pretty funny title.
Does anyone know why the covers were usually like this in the 70s? They are so characteristic and replicated across many books that I bet there is some back story<p>EDIT: wrong decade there
These are great and very surreal. I have always been a fan of the old sci-fi covers. These however look like something an artist might envision during a fever dream.
The thing with AI generated images is that often a lot of the patterns are repeated elsewhere in the image which to my eye always looks odd and feels “machiney”.
> The Past is Fictional<p>It’s interesting to think about how we tend to grade the “smarts” of machine intelligence by comparing them to what us humans believe to be the truth. Even if they were trying to teach us something how would we ever know?