Move AI continued to expand its markerless motion-capture system this year, and the pitch is quietly radical: extract clean, animation-ready 3D body motion from ordinary video — multiple cameras for the highest fidelity, or even a single phone for quick capture — with no motion-capture suit, no markers, and no dedicated volume. What used to require a specialized studio, tens of thousands of dollars of gear, and a technician now runs from footage a solo creator can shoot on hardware they already own, then export to standard animation and game pipelines.
Watch: Move AI’s Move One Motion Capture iPhone App Launch (YouTube)
Why markerless changes who gets to animate
Traditional optical mocap works by tracking dozens of reflective markers on a suit with a ring of expensive cameras — accurate, but gated behind cost and space that put it out of reach for individuals and small teams. Move AI replaces the markers and the rig with computer vision and learned human-motion models that infer the skeleton directly from normal images. The consequence is access: an indie game developer, an animator, a choreographer, or an artist can capture a real performance and drive a 3D character with it, on an indie budget. It’s the capture counterpart to the generative tools this site tracks — instead of synthesizing motion, it digitizes a real body performing.
Where it fits, and the limits
Exported motion drops into the standard animation pipeline — retargeted onto characters in Blender, game engines, or VFX tools — which is what makes it practical rather than a novelty. The tradeoffs are the honest ones for vision-based capture: heavy occlusion (limbs hidden behind the body or props), very fast or unusual motion, and precise finger detail remain harder than a high-end optical stage delivers, and multi-camera setups still beat single-phone capture for fidelity. But for the enormous middle of creators who were simply priced out of mocap entirely, “shoot a video, get usable animation” is a genuine shift in what’s possible.