Real-Time Marker-free Motion Capture from multiple cameras

Brice Michoud (1) , Erwan Guillou (1) , Hector Briceno Pulido (1) , Saida Bouakaz (1)

1 SAARA - Simulation, Analyse et Animation pour la Réalité Augmentée

We present a fully automated method for real-time and marker-free 3D human motion capture. The system computes the 3D shape of the filmed person from a synchronized camera set. We obtain a robust and real-time system by using both a fast 3D shape analysis and a skin segmentation algorithm for human tracking. A skeleton-based approach eases the shape analysis. We are able to track fast and complex human motion in very difficult cases, like selfocclusion. Results on long video sequences with rapid and complex movements, demonstrate our approach robustness.

Domaines: Informatique [cs]

Brice Michoud, Erwan Guillou, Hector Briceno Pulido, Saida Bouakaz. Real-Time Marker-free Motion Capture from multiple cameras. ICCV'2007 : Eleventh IEEE International Conference on Computer Vision, Oct 2007, Rio de Janeiro, Brazil. pp.1-7. ⟨hal-01502165⟩