I…what? Who thought this was a good idea, or even a thing that was needed? Why would you need to ‘modernize’ historical artwork? Why the great flying fuck would you do this by putting it through an AI program that - extremely fucking crucially - changes all of the minor details of the piece? This is absolutely terrible. Whoever worked on this needs to unplug all of their devices and go peel some potatoes.
Check out my milk bread posts if you’re into potatoes.
Can you share more about what information you think is lost in the high fidelity generations like the lighthouse and colossus, especially given you can toggle the images?
The fidelity with the lines and subject is pretty high right now via gpt4 vision and controlnet, but I could work on getting it higher – Which images bother you the most and what level of fidelity/auto restoration would cause you to have a more positive reaction?
Thanks for the feedback!
My problem is the basic concept. These are historical documents, and you are tampering with them. It’s like translating the Declaration of Independence into leetspeak. As a one-time gag, it’s worth a chuckle, but the idea that it would be an ‘improvement’ or a ‘modernisation’ is an insult. The ‘fidelity’ of your process is irrelevant. You do not and should not need an artifical ‘higher resolution image’ of a centuries-old painting.
That’s what I thought – It seems like there’s a group of people this bothers and aren’t interested in regardless of the outputs. It feels ideological, but I won’t go that far and claim that for you or them – Anyway, this isn’t made for you and it is OK that we feel differently.
Thanks for sharing your feedback and opinion again! Have a great day!
That wooly mammoth mangina is not an improvement on the original.
I’m going to lock “I’m old greegggggggg” into the generation prompt
Absolute garbage
Thanks for the feedback!
The first examples make my previous opinion change a bit. The first three kind of work: they are improvement over later depiction, they add a layer of speculation with more quality, why not. But you really need to be careful about proposing “improvements” of primary sources, see the comments you get here or on imgur. The fresco with 4 character is an example of what should not be done: you turn a roman fresco into a renaissance painting (looks like Rubens style?).
Your tool can be useful, you changed my mind about it with the lighthouse of Alexandria reconstruction. But really, choose your examples more carefully. Some are akin to writing a manga version of Batman and calling it an improvement. It is a core difference in style that not everyone will like.
Yes – I must never use the word ‘improve’ again that is clear haha – Do you like ‘modernize’ ‘update’ ? which words are least upsetting?
Somewhere else someone gave me the idea to build different fine tune models that are more aware of styles and techniques from different periods. Thanks for the great feedback I appreciate you!
I wanted to not cherry pick examples and so I just did 15 images and posted them to see how the ‘anger’ reaction has changed since last time.
I would suggest to maybe use it more on imaginary rendition, of fiction or literature. Or on colorizing some specific styles like engravings.
I actually have an old project that illustrates books automatically – https://github.com/pwillia7/booksplitter
I haven’t looked at it in a while – good idea and I’ll try to rebuild that on this stack. Here’s an example output from that (this is pretty ‘dumb’ and is just generating prompts based on the text and has some style locking with prompts) https://docs.google.com/document/d/1IsnynQZoxOBmZx9Jac4DfWn15YCevG63CxsIkbu8tgE/edit
Kinda interesting to see what AI can do but I really hope you didn’t actually upload these to Wikipedia
It’s a chrome extension that loads/saves them to an external DB and let’s you toggle between them
How is rewriting history an “improvement”? I fear the day when this will actually be uploaded to Wikipedia (be it in the open or secretly) and history will be erased by it.