# Mocap — Process
**Role in the pipeline:** body and face performance for monologues, gestures and idles.
## Capture options (Sept 2026)
| Option | Quality | Cost | Use |
|---|---|---|---|
| **MetaHuman Animator, single camera** (body + optional face, Windows) | Good | Lowest | Previs, gesture library, idles |
| iPhone TrueDepth / Live Link Face (iOS; Android app also records takes) | Good face | Low | Face for monologues and pickups |
| Stereo head-mounted camera + MetaHuman Animator | Best face | Medium | Hero monologue face |
| Inertial suit (Rokoko, Xsens class) | Good body | Medium | On-island capture weeks |
| Optical stage (Vicon/OptiTrack) rental | Best body | Highest per day | Hero monologue body |
## Steps
1. Script and blocking locked with historian and director.
2. Calibrate; slate each take; record reference video and clean audio with timecode.
3. Solve face in MetaHuman Animator; solve body (suit/stage or mono camera).
4. Clean in Maya or Unreal (feet, hands, props); retarget to the MetaHuman skeleton.
5. Build the **gesture library** tagged by rhetorical move (point, open hand, write, raise telescope).
6. Build **idles** (breath, weight shift, look-at) for kiosk rest state.
7. Bake compressed versions for phones and standalone headsets.
**Owners:** technical animator, mocap technician, director, actor. Tutorials: [[Mocap — Tutorials]]