Live Link Face

Live Link Face helps creators who need live facial performance capture for MetaHuman characters during previs, rehearsal, or on set. It uses your phone to track your face and send animation to a Windows based Unreal Engine machine in real time. Additionally, this beta targets compatible Android devices so you can trial the workflow with existing rigs. If direct device to engine streaming fits your setup, install the Live Link Face APK. It suits teams who need quick visual feedback while blocking scenes or testing rigs.

Live Link Face Features

Real time MetaHuman streaming

Live Link Face generates facial animation on your device and streams it over a network to Unreal Engine using the MetaHuman Live Link Plugin. You can apply the capture to any MetaHuman character in just a few clicks.

Timecode and multi device sync

For multi device shoots, you can align sessions with timecode options that match your stage setup.

• Use the device system clock
• Connect to an NTP server
• Use a Tentacle Sync device with a master clock on stage

Verified compatible phones

To help you choose hardware, the following models were verified in internal testing.

• OnePlus 12 5G
• ROG Phone 7 Ultimate
• Samsung S23 Ultra
• Samsung S24 USA, Canada, China
• Samsung S24 Ultra
• Sony Xperia 1 VI

Chipsets and setup requirements

Before configuring your pipeline, review chipset and platform notes.

• Realtime support targets Qualcomm Snapdragon 8 Gen 1, 8+ Gen 1, 8 Gen 2, 8 Gen 3, and 8 Elite.
• Compatibility is not guaranteed, as some manufacturers may disable required features.
• The Snapdragon 8s series is not supported.
• A desktop PC running Windows 10 or Windows 11 is required for the Unreal Engine workflow.

Start capturing on your device

This beta focuses on real time MetaHuman facial animation streamed from supported Android phones to Unreal Engine on Windows. Confirm your device and chipset, then set up the MetaHuman Live Link Plugin on your workstation. When ready, download Live Link Face and connect it to your project.