EXPERTIMENTAL ANY LORA SUPPORT
To get the proper LoRA use this workflow , then IGNORE THE WARNING in the prompt,
First Image is what you want to Animate,,
Automatic Description might be less useful than initially though, NONE recommended, but option is left open.
Second Image is your "Context" the closer to 1:1 the better, as this is fed to CLIP vision. You can use the same image, but it will be cropped to the center if it's not 1:1 so prompt adhesion might suffer.
v0.7.1
Increases the sizes to match recommended sizes more closely
V0.5.1
Adds ability to turn off auto-tagging.
Max video size is now 0.5 Mega Pixels or your choice of larger side, whichever is smaller. This tool is meant to be quick and economical allowing for more iteration, If there is demand I can make one for larger videos.