patch-gg / anymove

Motion capture for any creature from one AI video: two cameras in one frame, a DINOv2 joint reader, the rig fitted in both views, a GLB out.

MaxLiu22 / Motion2MixamoRig

Retarget human motion from video to an Adobe Mixamo-rigged 3D character.