Skip to content

Synthesizing Moving People with 3D Control

Animate a single image with target 3D motion

Visit official site

What Synthesizing Moving People with 3D Control is

Synthesizing Moving People with 3D Control is described as a diffusion framework that turns a single image into a person animation guided by a target 3D motion sequence.

What you can do with it

  • Render new body poses with clothing and hair filled in
  • Match an image sequence to 3D pose targets
  • Build synthetic camera paths around the person

Model notes

  • Priors cover hidden parts of the body and clothing
  • Training runs in texture map space
  • Output stays visually close to the input image

Pros & cons

✓ What we like

  • Image sequences stay visually similar to the input image
  • The method handles sustained motion and complex poses

! What to watch out for

  • Animation starts from a single input image
  • Output follows the given target 3D pose sequence

FAQ

What is the input?

The material describes one image plus a target 3D motion sequence.

What does 3D control add?

It allows image sequences matching 3D pose targets and synthetic camera trajectories.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review