Coming up update
Added 2024-01-14 17:07:13 +0000 UTC
From a messy diagram with lines flying around...
Now, we have a clear picture of everything, plus, I added back the background masked from version 7, and... something new of course.
Comments
Um, looks like you are new to ComfyUI and Diffusion AI model. First you need to understand What Is IPAdapter, then learn about the denoising in Stable Diffusion. I wouldn't have time to do summery on every version that it update. The AI is moving fast , and everything is keep testing. All characters and background is depends on the setting.
Benjamin Law
2024-01-21 13:05:13 +0000 UTCIt would be helpful if you could create and share a post with the artwork used for a video processing so we can replicate it in order to learn the process. I have watched your current and previous videos around video generation and I am still confused about a few things. For instance, I have an existing video for which I would like to change the look and feel of the person, remove the original background and insert a different background. I tried doing it a bunch of times without success (the person looks distorted and the background doesn't change much) and I have a feeling that I haven't understood some parts. Would appreciate a tiny summary for those who are late to the party. Many thanks!
John Smith
2024-01-21 12:22:03 +0000 UTCIn the above (video) example, is the background an image or a video? And do the resolution of the background and the foreground video need to match?
John Smith
2024-01-21 12:17:42 +0000 UTC
