Jump to content

AI generated art


Whitegemgames

Recommended Posts

TL;DR: AI art programs are now at the point where if you have the time and patience, you can create exactly what is in your head without skills in drawing. As of writing the most advanced ones are "Stable Diffusion", "Midjourney", and "Dall-E 2" with stable diffusion being the most customizable and free if you can run it on your own computer.

 

It's been a long time since I used this site and even longer since posting but I felt the need to share this. Since at least 2019 I have attempted off and on to use AI art generators to make images of Crystal with mixed results but some success here and there and often needing heavy editing. But within the past few months this tech has grown to an insane degree and for the first time I have been able to create an image we both agree is exactly how she looks in my head (other than the clothes but she is considering changing that). The Final result was made using "Stable Diffusion" and you can find plenty of guides and info on the subreddit, but the initial face was made on a site called "Waifu Lab", and the initial body was created on "Dall-E 2", but while it helps a lot to have a starting point everything could have been done using just stable diffusion and an image editor like photoshop or gimp. The best results take time, many versions of the images generated, and combining them all in an image editor, the final result is the combination of at least 20 maybe 30 images and that's not counting the ones I generated but didn't use. if you want it to be perfect you will need to put in the effort but if your more relaxed you could probably get something you like sooner.

 

The initial Face made using "Waifu Labs"

448285478_106Edit.png.df77d1e63caac6007a90d4f917fa0326.png

 

The Initial Body made using "Dall-E 2"

1764736421_DALLE2022-08-2120_22.48-Animegirlstandingonabalconyatsunset.thumb.png.f9552188981c868fbd15a8bf4fc4e344.png

 

First version using "Stable Diffusion"

795048488_12022-09-30-02-24-48-01-girl_standing_on_a_balcony_at_sunset_by_Tyler_Edlin-1632508170-scale7.50-k_euler_aedit2.thumb.png.e5e6b251c1521aca2d90a9ff745c430b.png

 

The final result using "Stable Diffusion", 8th version and all versions are multiple combined images.

2126637835_9Touchups2022-13.thumb.png.0c3c0fcca4b943b4e2b686c3231b633d.png

Edited by Whitegemgames
error loading final image

If text is green, Crystal is talking

If it's normal, its me

Link to comment
Share on other sites

That's pretty awesome!

I personally have had pretty much 0 luck with AI art of my tulpas.  More often than not, I end up with some grotesque abomination that ignores parts of my prompt and doesn't seem at all influenced by any input image I give it either.

I guess part of the problem is that none of my tulpas are human though.  The couple of times I've tried AI art of humans, it's worked decently well.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...