[synapticpaint]

It sounds like your app turns photos into a watercolor style image? You can definitely do this faster with Stable Diffusion (~6s per image before optimization).

Here's how I would approach it: train a dreambooth model on watercolor style images, then run image-to-image using that model.

For examples of what dreambooth models can do see: https://synapticpaint.com/dreambooth/info/ (sample images here generated using a "modern disney" style model).

If you need help getting this set up feel free to email me! This stuff is probably not harder than getting gimp to run in a container.

[wongarsu]

Stable Diffusion on beefy hardware is faster than OPs process in OPs docker container, but I don't think we can judge from this that OP uses a slower process. For all we know this is running on the equivalent of a $10 Digital Ocean droplet.

[nfriend]

Currently running on a t3.medium AWS EC2 instance! (My max budget is currently $50/month since I get that much in AWS credits each month for building an Alexa skill a few years back.)

[synapticpaint]

That's true. I just wanted to provide some context for OP and others reading since they mentioned an interest in AI generation, and I think not everyone knows that dreambooth + stable diffusion has this kind of capability.

[nfriend]

Very interesting, I will look into this! Thanks!

[rlv-dan]

This comment make me sad. Replacing human creativity with "ai" is like we've reached the peak so no need to put any effort into life anymore..

[synapticpaint]

I don't think that AI image generation in its current form replaces human creativity. You still need a human to guide the process, compose the image, judge the output, etc. What's being automated is the manual, traditional process of painting and drawing.

I see this technology as more analogous to hand writing books -> printing. I predict that more people, not less, will be involved in creative industries related to visual arts (design, film, illustration, animation, etc.) as a result of this technology becoming more accessible.

[JoeAltmaier]

That's skipping over all the skill and training needed to draw, well, what most of us are unable to draw.

We can all write words with a pencil. That's not the same as drawing an original character or scene with skill.

[UncleEntity]

The photography industry would like to have a word with you.

Or all the (now dead) people who were saying the same things about photography.

[JoeAltmaier]

Yeah and photographing art and claiming it is yours, is surely copyright infringement.

These disingenuous arguments are not taking the conversation forward.

[jasonjmcghee]

I don’t think that was the point parent comment was making. I took it as all the skills you used to need for photography, like developing photos. Now you can point and click with a phone.

But everything else that goes into a good photograph, like what to use as your subject, the composition, post processing, etc is still important.

[JoeAltmaier]

Yes, this is different from much of art history. Instead of coming up with original characters or styles or backgrounds and palettes, we just ask a computer to do it all and usually by copying somebody that can do those things.

Art will take a hit to be sure, and the brunt of it will be taken by artists with actual skills.

[conductr]

It actually kind of excites me to see what kind of human creativity will blossom in the wake of AI. What art will humans eventually create that is so unique that it's obviously not AI? That's the exciting part, but it could be a difficult/long journey.

[jamesgreenleaf]

Human creativity isn't being replaced; new layers of it are being added.

[adammarples]

running a GIMP script is hardly the opposite