🏃 Motion Capture Without Suits or Markers
MAMMA (Markerless Accurate Multi-person Motion Acquisition) has been introduced—a pipeline for capturing the motion of multiple people from video. The system reconstructs SMPL-X model parameters via the MammaNet transformer network, even when using ordinary smartphones.
🌍 This is a significant step toward the democratization of MoCap technologies. Accurate capture of human interaction without expensive equipment (such as Vicon/OptiTrack) opens new possibilities for film, games, and metaverses.
👤 Sensors are no longer required to create 3D animation. Smartphones placed around a scene can be used to obtain precise data for Blender, Maya, or Unreal Engine.
Source 1: https://mamma.is.tue.mpg.de/ Source 2: https://github.com/cuevhv/mamma