You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, you can use DWPose to extract the skeleton pose sequence from the driven video before initiating StableAnimator. For more details on skeleton pose extraction, please refer to the README file.
I think you mean to say instead of reference.png used reference.mp4 which I think is not going to work. What you can do is extract the first frame from input video and save it as reference.png and use it.
It seems I misunderstood the question. This setting can be categorized as video motion editing. For more details, you can refer to MotionEditor or MotionFollower.
Can video be used as input? Don't change the faces of the people in the video, just the actions?
The text was updated successfully, but these errors were encountered: