latentblending VS course22p2

So it ought to be pretty easy to implement latent blending on source and target images which were not generated by SD?

latent blending: https://github.com/lunarring/latentblending This one has a lot of potential, since it can be used to transition from one prompt to the next. It interpolates between the prompst and also the image latents themselves.

you can stay tuned also on twitter https://twitter.com/j_stelzer or github https://github.com/lunarring/latentblending/

yip, long interpolation using experimental version of latent blending (https://github.com/lunarring/latentblending/). you will be able to do this yourself very soon with a nice gradio UI.

Github: https://github.com/lunarring/latentblending/

Generated some images using that model and picked ones that had similar composition to create the videos using latent blending: https://github.com/lunarring/latentblending/

Latent blending is a amazing, and produces effects quite different to the available extensions. https://github.com/lunarring/latentblending

Practical Deep Learning for Coders: https://course.fast.ai/Lessons/part2.html

To me the most important bit is that the diffusion loop turns a noisy latent into an image, does that iteratively, and uses "guidance" in the form of a prompt/controlnet image/etc to do it. The scheduler part, I felt, was needlessly complex for this short explainer, so I hand-wave it away. IF someone wants to dive in deeper, much deeper, they can go through the same thing I'm doing, which is this: https://course.fast.ai/Lessons/part2.html

They also recently released a course for more experienced students where they teach you to implement the Stable Diffusion algorithm from scratch.

The full transcripts are available here in plain text form:

https://github.com/fastai/course22p2/tree/master/summaries