SteadyDancer_ Video Action Transfer - Image to Video - One to All-Human Bones
Details
Download Files
Model description
Bilibili: AIGC特异点
Youtube： https://www.youtube.com/ @AIGC-Singularity
You can click on the link below to try it out directly. If the effect is good, you can deploy it locally
https://www.runninghub.ai/post/1999132133182631937/?inviteCode=sdhs0trb
Fan benefits,register to get 1000 points,daily login 100 points,play 4090! Experience the super power of 48G.
Video Generation Duration Settings:
For a 5-second video, use the original default settings.
To generate a 10-second video, the required parameter range must be adjusted from $150$ to $300$.
Core Node: Pose Detection OneToAll Animation
This workflow's most critical component is the Pose Detection OneToAll Animation node. Within its alien_to configuration, you can select between two alignment modes: ref and pose.
OptionAlignment DescriptionBest Use Case****RefScales and aligns the skeleton image to the reference image.Cartoon characters and stylized dancing videos.PoseScales and aligns the reference image to the skeleton image.Videos featuring a normal human body shape or realistic dance.
Model Download:
Link: https://pan.quark.cn/s/0c19d1452a6c
Node download address
Link: https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/onetoall