Recent content by ripptechlabs

Hi and welcome! So the answer is... maybe. If you need it upscaled and everything *exactly* as it is but just bigger, you're going to have to deal with photoshop etc. Just resize and set the DPI. It might not be brilliant but it will be better. And its still your exact image. If you...

1. You can usually input at least one image to use as a "reference" ... I know MidJourney has this, pretty sure Stable Diff allows for it. I can't speak to the rest as I don't use much else. 2. Again, Midjourney just introduced the 'character reference' --cref option. You could probably use...

you're not wrong, and it seems to be getting worse. addressing this *is* on my list but its not near the top atm. Stay tuned.

version 6 is getting really really good.

have you tried getting a closer-up shot of the girls facing away, and then using expand to fill in the rest of the image? I honestly don't know if it would work like you want or not. Just the first thing came to mind looking at these.

feed it back into MJ or DALL-E asking for a prompt (or "/describe" I think in MJ) and see if it will give you something close. Season to taste.

not really, I don't think. You'd need to AI upscale them and then change the DPI in another application. If anyone knows a generator (Stable Diff?) that allows you to set the DPI of the output PNG chime in!

inpainting. Its not quite as simple as what you're looking for, but close. You paint a mask over the part you want to change, and then give the prompt for that. You might not be able to change some aspects like "make it rainy scenery" without it re-writing the whole image. But smaller stuff...

I'm not 100% on this, but I think Stable diffusion can be set up to do this by training an addon model with all of your subject's images. I saw a video on YouTube but can't recall atm who it was or when. I'll see what I can find and report back. I still think it gave a best approximation and...

huh, that's weird. what's weirder is I tried this in Midjourney, bc why not... and the results aren't all that different, check it out. Same sort of looking down at it. I still think its strange that there are some things that it just can or will not do, and other things that you can...

just for S&G I figured I'd spin over to dalle and see how things were progressing. my prompt: the results? not. too. bad. They've souped things up since I last played with it. I've been b*** deep in MJ and GPT-4 trying to launch some stuff and haven't paid as close attention as I should...

I heard about this on Matt Wolfe's(?) YT channel. was like 😲😲

(just for fun ima throw these at GPT-4 and see what it says lol) ========= gotta love v4 and its lists

Hey and welcome to the site! I'm not too versed in Leonardo, or Stable Diffusion, but let me take a crack at it. 1) Yes. It will use your image as a starting point, but you still need to tell the AI what to do with it. Otherwise it really doesn't know. Maybe there's a way to use img2img...

Also I took the first prompt and shoved it into Leonardo just to see what would happen. Not sure if I like it or not tbh.