DAINAPP patreon

Outpainting interface for SD

🕑 Added 2022-10-03 18:34:33 +0000 UTC

Comments

Rart

looks great ! im not sure is it possible to create video output now Gui ? like this https://www.youtube.com/watch?v=_074tePxU3k

STR4NGΞR D4NGΞR

Oh MAN this needs to happen. I've been messing with the current (.52?) video img2img method and although interesting, nothing release-quality yet. I would LOVE to see VToonify or some equivalent implemented!

GRisk

That is a lot of stuff combined. Not sure if making a GUI for everything will be a good idea, maybe in a far away future.

VToonify, which allows you to seamlessly apply a style to any video you want via images generated by Stable Diffusion and learning, looks interesting, and I would be happy to see it as a GRisk feature. https://twitter.com/ShuaiYang1991/status/1577251207155838976 https://huggingface.co/spaces/PKUWilliamYang/VToonify

This is great news! Thank you!

Oof YES Keen as dude, thats absolute pog

this is the main feature I am waiting on

Great to hear!

wow nice!

GRisk

For now it auto adjust using the selected resolution, as show in the gif. In time more tools will appear.

GRisk

Thanks!

GRisk

No problem. Models that don't use Diffusers new a special code. I'm working to add it to the GUI

GRisk

Other schedulers are in my list to add

I'm trying to use .ckpt models from https://rentry.org/sdmodels#stable-diffusion-models and I get this Render Start Loading model (May take a little time.) Loading model from file gg1342_testrun1_pruned.ckpt Traceback (most recent call last): File "start.py", line 944, in OnRender File "start.py", line 845, in LoadModel AttributeError: 'dict' object has no attribute 'enable_attention_slicing' also sorry for commenting this in two places I just wanna make sure it gets attention