MultiDiffusion VS auto-sd-paint-ext

For composing this approach works pretty well

https://multidiffusion.github.io/

So I've continued to experiment with how many papers I can fit into a single pipe and have them play nicely together. The images below were created by combining the panorama code from omerbt/MultiDiffusion with the ideas from albarji/mixture-of-diffusers. Also turns out nateraw/stable-diffusion-videos can be seen as a special case of a panorama (in latent space rather than prompt space).

Essentially, I fine-tuned Stable Diffusion 2.1 base (the 512x512) model on the ESA Hubble Deep Space Images & Captions dataset I collected from public Hubble images & captions. After around 33,000 training steps, I saved the model and was really impressed by the results. But I really wanted to be able to generate wallpaper-level quality space images, so I stumbled upon MultiDiffusion: a new project for generating massive panorama images using stable diffusion models. I then used hubble-diffusion-2 along with MultiDiffusion to generate each one of these amazing 2560x1536 images. Each image took a little over an hour to generate on a Google Colab T4 GPU. I used the following prompts for each of these images:

If you need generated and not upscaled 4k for some reason, try something like https://github.com/omerbt/MultiDiffusion

Project: https://multidiffusion.github.io/ Paper: https://arxiv.org/abs/2302.08113 GitHub: https://github.com/omerbt/MultiDiffusion

Quick Read: https://www.marktechpost.com/2023/02/24/meet-multidiffusion-a-unified-ai-framework-that-enables-versatile-and-controllable-image-generation-using-a-pre-trained-text-to-image-diffusion-model/ Paper: https://arxiv.org/abs/2302.08113 Github: https://github.com/omerbt/MultiDiffusion Project: https://multidiffusion.github.io/

So I haven't made many images with Stable Diffusion despite using it heavily. The reason is I've been messing with the internals of the diffusion pipe, to interfere with the diffusion process in different ways. Todays fun result is based on omerbt/MultiDiffusion for making panoramas.

If we combine Mixture of Diffusers + MultiDiffusion+ Composer+ cross-domain-compositing and probably some more I'm not thinking of.

It makes more sense to embed stable diffusion capabilities into well-established image editors such as Gimp, Photoshop, Krita, or Figma, which come with layered, non-destructive functionalities, rather than attempting the opposite approach.

https://github.com/Interpause/auto-sd-paint-ext

You might be interested in knowing that Krita has a SD plug in that I've found works quite well if you're willing and able to install and run SD locally. Krita is now my primary art program. https://github.com/Interpause/auto-sd-paint-ext

GitHub - Interpause/auto-sd-paint-ext: Extension for AUTOMATIC1111 to add custom backend API for Krita Plugin & more

Yep, Krita has a SD plugin too ( https://github.com/Interpause/auto-sd-paint-ext ) but probably not as good as this Photoshop one yet.

Again in my humble opinion, however, it's nice to be able to choose (at least if you're willing to pay the subscription costs) among different models and algorithms. Use whatever suits best your needs or your ideas. I personally prefer SD's versatility, and the workflows enabled by "being able to get my hands on all the parameters", like sampler selection, CFG scale, inpainting and other stuff (have you tried InvokeAI's unified canvas or Automatic1111's Krita plugin?). Not to mention the ability to do progressive and potentially infinite upscaling of an image (Automatic1111's "SD upscale" script or InvokeAI's "Embiggen").

I have a simple question that apparently i cant seem to figure out for the past like 12 hours, im using the https://github.com/Interpause/auto-sd-paint-ext auto extension for the krita plugin, and im simply just trying to inpaint creatures and just general things into a scene, but no matter what i do, the result is nothing like it, i followed the inpainting tutorial they have listed on github (New layer from visible) then make a new layer to paint the mask on, tried all fill methods with different steps, im fairly new to all of this and what not, im using a custom model checkpoint btw if that matters (protogen 5.8)