This is a collection of links for video generation with Stable Diffusion. This article covers methods using ComfyUI.
AnimateDiff
A model that adds layers to Stable Diffusion so that it can learn the time axis.
>-
Stable Video Diffusion
Currently, it only supports image2video, but the video accuracy is good (there is no flickering like AnimateDiff).
>-
AnimateDiff-Lightning
A lightweight model that enables video generation in a few steps by increasing the noise removal efficiency of AnimateDiff. You can generate with good accuracy even in 2-3 steps.
>-
RIFE
A model that performs frame interpolation of videos (increases frame rate). It is possible to convert to a smooth video without degrading the quality of the video.
>-
vid2vid
Various methods are used for vid2vid depending on the person, but here is an example of what I generated.
>-
