4DAnyone: Create 4D human avatars from casual monocular video

rsasaki0109 · x · 2026-08-21

Accepted to SIGGRAPH Asia 2026, 4DAnyone turns a casual monocular video into multi-view videos for downstream 4DGS reconstruction. It outputs a 4D human from a single video input without the need for rigs, calibration, or tripods.

Related event: 4DAnyone Generates 4D Human Avatars from Monocular Video(3 posts)→

Original post →

More from Embodied

Embodied channel →