Stories
Slash Boxes
Comments

SoylentNews is people

Unreal’s New iPhone App Does Live Motion Capture with Face ID Sensors

posted by martyb on Saturday July 11 2020, @06:12AM   Printer-friendly
from the when-you-need-mo'-mo-cap dept.

upstart writes in with an IRC submission:

Unreal’s new iPhone app does live motion capture with Face ID sensors:

Unreal Engine developer Epic Games has released Live Link Face, an iPhone app that uses the front-facing 3D sensors in the phone to do live motion capture for facial animations in 3D projects like video games, animations, or films.

The app uses tools from Apple's ARKit framework and the iPhone's TrueDepth sensor array to stream live motion capture from an actor looking at the phone to 3D characters in Unreal Engine running on a nearby workstation. It captures facial expressions as well as head and neck rotation.

Live Link Face can stream to multiple machines at once, and "robust timecode support and precise frame accuracy enable seamless synchronization with other stage components like cameras and body motion capture," according to Epic's blog post announcing the app. Users get a CSV of raw blendshape data and an MOV from the phone's front-facing video camera, with timecodes.

[...] For those not familiar, Unreal Engine began life as a graphics engine used by triple-A video game studios for titles like Gears of War and Mass Effect, and it has evolved over time to be used by indies and in other situations like filmmaking, architecture, and design. It competes with another popular engine called Unity, as well as in-house tools developed by various studios.

Original Submission

 
This discussion has been archived. No new comments can be posted.
Unreal’s New iPhone App Does Live Motion Capture with Face ID Sensors | Log In/Create an Account | Top | 3 comments | Search Discussion
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

  • (Score: 2) by FatPhil on Saturday July 11 2020, @07:37AM (1 child)

    by FatPhil (863) <{pc-soylent} {at} {asdf.fi}> on Saturday July 11 2020, @07:37AM (#1019412) Homepage
    How is this different? https://learn.g2.com/animoji

    It looks like it's just an animoji in a game context rather than a messaging context. How is that advancing the field? Next innovation - embed it in the music playing app so that you san see yourself sing in 3D along with the music? I'm a geniuose, I should patent that.
    --
    Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
    Starting Score:    1  point
    Karma-Bonus Modifier   +1  
    Total Score:   2  

  • (Score: 2) by c0lo on Saturday July 11 2020, @07:56AM

    by c0lo (156) Subscriber Badge on Saturday July 11 2020, @07:56AM (#1019418) Journal

    My understanding, so I may be totally wrong

    The difference is that the output of animoji is just an animated pile of poo (that maybe mimics your facial expression) while this one streams live motion capture data.
    I assume animating a pile of poo requires less samples for mimicking something good enough, but one expects a lot more to have something realistic.

    The fact that one use Unreals engine to morph MoCap into the face of one's character of choice is incidental.

    --
    https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford