stable-diffusion VS txt2imghd

Yes, you can install it with Python! https://github.com/lstein/stable-diffusion works with macOS, and you can control all the common parameter via their WebUI or CLI :)

I'm using lstein fork ("dream") and when I create an image from the terminal, it also writes back to the terminal like this:

Stable Diffusion is wild - the space has been quickly developing and watching the pace of development makes me reconsider what I consider "staggering". I've been blown away. The accessibility of this technology is even more incredible - there's even a fork that is working on M1 Macs (https://github.com/lstein/stable-diffusion)

We are in for some interesting times. Whatever the next iteration of Textual Inversion is will be extremely disruptive, especially if the concepts continue to be developed collectively.

Oh, it generates from a text prompt, not a sample texture. I thought this was just a tool to generate wrapped textures from non-wrapped ones.

The licensing is a mess. The Blender plug-in is GPL 3, the stable diffusion code is MIT, and the weights for the model have a very restrictive custom license.[1] Whether the weights, which are program-generated, are copyrightable is a serious legal question.

[1] https://github.com/lstein/stable-diffusion/blob/61f46cac31b5...

> Whenever I ask for something like ‘seamless tiling xxxxxx’ it kinda sorta gets the idea, but the resulting texture doesn’t quite tile right.

Getting seamless tiling requires more than just have "seamless tiling" in the prompt. It also depends on if the fork you're using has that feature at all.

https://github.com/lstein/stable-diffusion has the feature, but you need to pass it outside the prompt. So if you use the `dream.py` prompt cli, you can pass it `"Hats on the ground" --seamless` and it should be perfectly tilable.

From https://github.com/lstein/stable-diffusion

Yes, works perfectly fine as of 1e8e5245ebca5211e271f35a3a849dee8f4793d2 which contains the performance improvements. Probably works fine with later commits too but I haven't personally tested them so won't vouch for it.

The dictionary definition of upscale is to increase the size or improve the quality of. or convert (an image or video) so that it displays correctly in a higher resolution format. It's just scaling in one direction, the opposite of downscaling. Even the original author of high res fix uses word upscaling because the original library uses Real-ESRGAN to upscale transient images. A1111 repo provides an option to use an upscaling model or just bilinear interpolation for this step.

(4) What the SD upscaler does is chop the image into tiles, re-run SD on the tiles, and then stitch them back together. img2imgHD, and this custom script are other options that do a similar thing. You can also do it manually and combine the tiles in photoshop/gimp/krita, but it's a lot more work.

https://github.com/jquesnelle/txt2imghd The link was literally in the Auto's code's comments.

This exact thing DOES exist... kinda. https://github.com/jquesnelle/txt2imghd Not sure if there's a Colab notebook for it yet, though. I'd use it in a heartbeat if there was one.

I was looking for a new script that could implement this process directly with the prompt and I found this: https://github.com/jquesnelle/txt2imghd/blob/master/README.md