Page 3 of 4

Re: Foreverlurk's AI works

Posted: Thu Oct 19, 2023 10:54 pm
by Hand-Holder
THE FUTURE IS BRIGHT !!!!!!!!!!!!!!!!!!!!!!!!!!!!111

Re: Foreverlurk's AI works

Posted: Wed Oct 25, 2023 9:35 pm
by foreverlurk
Hello again!

Just wanted to share a few things I posted this week. I'm playing around with different styles, sizes, facial expressions, angles... etc. Hope you like them.

Image

Image

Image

Image

Re: Foreverlurk's AI works

Posted: Wed Oct 25, 2023 11:20 pm
by bobvan2
Cool what about the 1 with the shrink ray

Re: Foreverlurk's AI works

Posted: Wed Oct 25, 2023 11:35 pm
by foreverlurk
bobvan2 wrote:
Wed Oct 25, 2023 11:20 pm
Cool what about the 1 with the shrink ray
Not sure what's your question, but if you mean it looks different, you're right. It's a more complex scene I created with the DALL-E 3 Beta in ChatGPT. Contrary to the free Bing Image Creator, it allows to create persistent characters (to a certain degree), you can then tweak the same scene, changing angles, lighting, clothes (again, to a limit).

Re: Foreverlurk's AI works

Posted: Thu Oct 26, 2023 12:45 am
by Pacific1982
DocRick wrote:
Wed Oct 04, 2023 11:12 pm
Here's a few stock images from my library that might help.


Image

Re: Foreverlurk's AI works

Posted: Thu Oct 26, 2023 1:54 am
by DocRick
Pacific1982 wrote:
Thu Oct 26, 2023 12:45 am
DocRick wrote:
Wed Oct 04, 2023 11:12 pm
Here's a few stock images from my library that might help.

Is that AI at all? I have to admit I know very little about AI, but I was under the impression he was looking for images to prompt the computer ????? This one looks simply like your old school beautiful photo manipulation techniques.

Re: Foreverlurk's AI works

Posted: Thu Oct 26, 2023 2:08 am
by Pacific1982
DocRick wrote:
Thu Oct 26, 2023 1:54 am
Is that AI at all? I have to admit I know very little about AI, but I was under the impression he was looking for images to prompt the computer ????? This one looks simply like your old school beautiful photo manipulation techniques.
It is old school beautiful photo manipulation. That’s how I did it. I suppose I misunderstood the idea that was being presented. I apologize for the misunderstanding.

Re: Foreverlurk's AI works

Posted: Thu Oct 26, 2023 4:10 am
by bobvan2
foreverlurk wrote:
Wed Oct 25, 2023 11:35 pm
bobvan2 wrote:
Wed Oct 25, 2023 11:20 pm
Cool what about the 1 with the shrink ray
Not sure what's your question, but if you mean it looks different, you're right. It's a more complex scene I created with the DALL-E 3 Beta in ChatGPT. Contrary to the free Bing Image Creator, it allows to create persistent characters (to a certain degree), you can then tweak the same scene, changing angles, lighting, clothes (again, to a limit).
I saw your image with a GT chasing and shrinking with SW with a ray gun or device..

Re: Foreverlurk's AI works

Posted: Sun Nov 05, 2023 4:14 pm
by foreverlurk
Some more recent ones... I actually bit the bullet and got the paid tier of DALL-E. It has pros and cons compared to the free tier, but the language model is way more advanced (the filtering too, sadly).

Image

Image

Image

Image

Image

Image

Image

Re: Foreverlurk's AI works

Posted: Sun Nov 05, 2023 8:56 pm
by rscholar
I think I officially have more of your stuff in my DA favorites than any single artist. You've really have a finesse for this.

Curious thing I found is that, despite how much I really like the tender sweet scenarios (seriously, you have a poet's soul), this one is one of my all time favorites:
foreverlurk wrote:
Wed Oct 25, 2023 9:35 pm

Image
I actually have a story idea for this if you don't mind - a youtuber pranks his wife by incrementally shrinking her without her noticing. When she finally sees it, she tries to confront him but he puts the device on full blast and she has to run to him angrily squeaking.

Re: Foreverlurk's AI works

Posted: Sun Nov 05, 2023 8:57 pm
by eddiegiantman
Is there any doll play related plans soon?

Re: Foreverlurk's AI works

Posted: Sun Nov 05, 2023 9:56 pm
by Hand-Holder
The Piano one is BEAUTIFULLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL

Re: Foreverlurk's AI works

Posted: Sun Nov 05, 2023 10:09 pm
by foreverlurk
rscholar wrote:
Sun Nov 05, 2023 8:56 pm
I think I officially have more of your stuff in my DA favorites than any single artist. You've really have a finesse for this.
Curious thing I found is that, despite how much I really like the tender sweet scenarios (seriously, you have a poet's soul), this one is one of my all time favorites
Oh wow - that's incredible praise, I don't know what to say. Thank you sincerely.

I honestly don't know why this fantasy inspires so much... Sehnsucht in me? Maybe I'm being overly dramatic and sentimental, too! Wouldn't be the first time, and it gets in the way of my darker, more dubcon scenarios. Even during RP I struggle to play the "evil" giant, which is weird because I also like it very, very much.
rscholar wrote:
Sun Nov 05, 2023 8:56 pm
I actually have a story idea for this if you don't mind - a youtuber pranks his wife by incrementally shrinking her without her noticing. When she finally sees it, she tries to confront him but he puts the device on full blast and she has to run to him angrily squeaking.
Yes please do! I'd be honored that you would choose to make a story from one of my pictures. That does sound a bit like what I had in mind actually! :D

Re: Foreverlurk's AI works

Posted: Sun Nov 05, 2023 10:14 pm
by foreverlurk
eddiegiantman wrote:
Sun Nov 05, 2023 8:57 pm
Is there any doll play related plans soon?
Depends, I always have plans but I have to make the AI gods do my bidding, just have to get lucky sometimes.
Hand-Holder wrote:
Sun Nov 05, 2023 9:56 pm
The Piano one is BEAUTIFULLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL
Thank you, really like that one. It's based on daydreams I had while doing piano lessons growing up (I got distracted sometimes hehe).

Re: Foreverlurk's AI works

Posted: Sat Nov 11, 2023 3:45 pm
by Pacific1982
Here is what I have to say about the AI generated images here. Despite any flaws—the lighting, shadows, and color matching are all spot-on.

Re: Foreverlurk's AI works

Posted: Sat Nov 11, 2023 10:25 pm
by foreverlurk
Pacific1982 wrote:
Sat Nov 11, 2023 3:45 pm
Here is what I have to say about the AI generated images here. Despite any flaws—the lighting, shadows, and color matching are all spot-on.
Thank you, I'm a fan of your work so it means a lot. I usually try to fix most mistakes and imperfections but it's not always possible, so I learned to live with them (this is eerily meta, but yeah).

Re: Foreverlurk's AI works

Posted: Sun Nov 12, 2023 3:37 am
by xrattletrapx
foreverlurk, these are fantastic! My favourite has to be the dungeons and dragons one! How in the hell did you prompt that with Bing?

Re: Foreverlurk's AI works

Posted: Sun Nov 12, 2023 8:24 pm
by foreverlurk
xrattletrapx wrote:
Sun Nov 12, 2023 3:37 am
foreverlurk, these are fantastic! My favourite has to be the dungeons and dragons one! How in the hell did you prompt that with Bing?
Thanks a lot! If you mean the ones with the D&D table and minis, they weren't done with Bing, but through ChatGPT. It's confusing I know, but there are four (4) ways to use DALL-E:

* Bing Image Creator
* Microsoft Copilot
* ChatGPT ($)
* OpenAI DALL-E API ($)

Each one has a slightly different flitering set, capabilities, limits, etc. I'm using the first three, and while I do have an API key, I haven't used it yet. ChatGPT helped me get the basic prompt and gens. Then I used Stable Diffusion to clean up the low res and garbled models. I have generated tons more like this one:

Image

Re: Foreverlurk's AI works

Posted: Tue Dec 12, 2023 4:10 am
by foreverlurk
Just some recent images I posted to DA, again a mix of DALL-E and Stable Diffusion.

All I Want For Christmas
Image

Lick Her Clean
Image

Rivalry
Image

Unbroken
Image

Re: Foreverlurk's AI works

Posted: Sun Jan 14, 2024 6:30 pm
by foreverlurk
Hello there!

Just posted some new gens on DA, here's a selection for you to enjoy.

Image

Image

Image

Image

Re: Foreverlurk's AI works

Posted: Sun Jan 14, 2024 6:39 pm
by who renders 3d
Fantastic! I love the two goth girls. But i guess some with a foot fetish combined with a mathematics fetish will get nightmares from the girl on the desk

Re: Foreverlurk's AI works

Posted: Sun Jan 14, 2024 6:45 pm
by foreverlurk
who renders 3d wrote:
Sun Jan 14, 2024 6:39 pm
Fantastic! I love the two goth girls. But i guess some with a foot fetish combined with a mathematics fetish will get nightmares from the girl on the desk
What, you don't have three "minus" operator on your calculator? :lol:

Re: Foreverlurk's AI works

Posted: Fri Jan 19, 2024 6:08 pm
by Little Sally
I was having a play with that free image creator from Microsoft Bing (doing the tentacle search I mentioned on another thread), and I came across another site which had an option for 'reimagining' pics you've already got, and I was quite impressed with the results. Trouble was I didn't notice it only gave you 5 free tries before asking for a monthly payment (or yearly). But it did give quality results. Anyway I moved on from it thinking I'd find some more that had this reimagine option, and now I've forgotten which one it was! Duh!

I did come across another free one called Stable Diffusion (I think), but the results on that were horrible with a lot of deformity of hands and other stuff.

Do you have any recommendations you might think worth trying?

Re: Foreverlurk's AI works

Posted: Fri Jan 19, 2024 8:06 pm
by foreverlurk
Little Sally wrote:
Fri Jan 19, 2024 6:08 pm
Anyway I moved on from it thinking I'd find some more that had this reimagine option, and now I've forgotten which one it was! Duh!
Is it possible you were editing the generated pics in Microsoft Designer? It's yet another way to access DALL-E:
https://designer.microsoft.com/image-creator

It allows to erase objects, remove or blur backgrounds, etc. I don't use it since I'm lucky enough to have a powerful graphics card to do all of this locally.
Little Sally wrote:
Fri Jan 19, 2024 6:08 pm
Do you have any recommendations you might think worth trying?
Like I said I use a local Stable Diffusion setup (https://github.com/lllyasviel/Fooocus) for "reimagining" pictures by inpainting the details in the existing pic. Online models aren't very good compared to fully tweaked ones with LORAs (and no censorship).

Otherwise, I also tried MidJourney and it's simply not as good for sizey content (but for landscapes and portraits it's better than DALL-E)

Re: Foreverlurk's AI works

Posted: Fri Jan 19, 2024 10:18 pm
by Little Sally
foreverlurk wrote:
Fri Jan 19, 2024 8:06 pm
Little Sally wrote:
Fri Jan 19, 2024 6:08 pm
Anyway I moved on from it thinking I'd find some more that had this reimagine option, and now I've forgotten which one it was! Duh!
Is it possible you were editing the generated pics in Microsoft Designer? It's yet another way to access DALL-E:
https://designer.microsoft.com/image-creator

It allows to erase objects, remove or blur backgrounds, etc. I don't use it since I'm lucky enough to have a powerful graphics card to do all of this locally.
Little Sally wrote:
Fri Jan 19, 2024 6:08 pm
Do you have any recommendations you might think worth trying?
Like I said I use a local Stable Diffusion setup (https://github.com/lllyasviel/Fooocus) for "reimagining" pictures by inpainting the details in the existing pic. Online models aren't very good compared to fully tweaked ones with LORAs (and no censorship).

Otherwise, I also tried MidJourney and it's simply not as good for sizey content (but for landscapes and portraits it's better than DALL-E)
Thank you, I'll look at these.

The one I mentioned that I first tried and liked, had the option where you could upload a pic (including work of your own) and AI would do interpretations of it. Or reimagining it as they like to term it.