The experience of creating my first video (Shigure Ui dance)

Software Used: 

1. Stable Diffusion (extensions: ADetailer Tiled Diffusion ControlNet (lineart))

2. After Effects

   
   It's time for the first post, these are my attempts to make an AI video.
   See the screenshots attached for detailed settings. I decided to make a small number of changes and to use the most standard method “Image to Image” for each frame.
   
   Many tests were then carried out to select the model that best suited the style.
   I used ADetailer extensively for facial detailing.
   Used Tiled Diffusion for image scaling.
   
   And of course, used ControlNet for geometry manipulation.
   
   Once I was pleased with the result, I went to rest for a couple of hours and everything was ready :) Since the source had a green screen, I decided to make the background using AnimateDiff and combined everything in AfterEffects.