NovelAI/Stable Diffusion Image Generation (New Images, Am I getting better?)

VegitoHlove

VegitoHlove

Member
Apr 27, 2018
358
895
So I got to play with the NovelAI Image generation utility. Making some anime gals. Some turned out better than others.The following images were made by feed text or rather tags, you can also use image reference on an image you upload. to get perfect images you'll have to play around with both the tags and the setting which dictate how an image is generated, you can then enhance, edit or generate variations on the same generated images. What do you guys think? Is AI try the same thing it does every night? Try to take over the world.

Let me know your thoughts.
You don't have permission to view the spoiler content. Log in or register now.



You don't have permission to view the spoiler content. Log in or register now.
You don't have permission to view the spoiler content. Log in or register now.

Just a note I am no longer using Novel AI at all as I no longer access it.
 
Last edited:
VegitoHlove

VegitoHlove

Member
Apr 27, 2018
358
895
OsamiWorks said:
learn more so you at least get the same character from image to image
Click to expand...
Was my first try. Wasn't trying to get the same character, I wanted to see how vague and also how specific I could be with my text prompts. I still have much to show after I've made some more progress.

You can get consistent character by using image reference, then using the settings to dictate how much freedom it has to change the image and how much detail it can add. Though it's more work than you'd think to get both high quality and consistent images. Still playing around with it.
 
You must log in or register to reply here.
Top Bottom