Brice Michoud

Real-Time Marker-free Motion Capture from multiple cameras

We present a fully-automated method for real-time and marker-free 3D human motion capture. The system computes the 3D shape of the person filmed 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 facilitates the shape analysis. We are able to track fast and complex human motion in very difficult cases, like self-occlusion. Results on long video sequences with rapid and complex movements, demonstrate our approach robustness.

Real-time and markerless 3D human motion capture using multiple views

We present a fully automated system for real-time markerless 3D human motion capture. Our approach, based on fast algorithms, uses simple techniques and requires low-cost devices. Using input from multiple calibrated webcams, an extended Shape-From-Silhouette algorithm reconstructs the person in real-time. Fast 3D shape and 3D skin parts analysis provide a robust and real-time system for human full-body tracking. Animation skeleton and simple morphological constraints make easier the motion capture process. Thanks to fast and simple algorithms and low-cost cameras, our system is perfectly apt for home entertainment device. Results on real video sequences with complicated motions demonstrate the robustness of the approach.

Real-time and marker-free 3D motion capture for home entertainment oriented applications

We present an automated system for real-time marker-free motion capture from two calibrated webcams. For fast 3D shape and skin reconstructions, we extend Shape-From-Silhouette algorithms. The motion capture system is based on simple and fast heuristics to increase the efficiency. Multi-modal scheme using both shape and skin-parts analysis, temporal coherence, and human anthropometric constraints are adopted to increase the robustness. Thanks to fast algorithms, low-cost cameras and the fact that the system runs on a single computer, our system is perfectly suitable for home entertainment device. Results on real video sequences demonstrate our approach efficiency.

Fast environment extraction for lighting and occlusion of virtual objects in real scenes

Augmented reality aims to insert virtual objects in real scenes. In order to obtain a coherent and realistic integration, these objects have to be relighted according to their positions and real light conditions. They also have to deal with occlusion by nearest parts of the real scene. To achieve this, we have to extract photometry and geometry from the real scene. In this paper, we adapt high dynamic range reconstruction and depth estimation methods to deal with real-time constraint and consumer devices. We present their limitations along with significant parameters influencing computing time and image quality. We tune these parameters to accelerate computation and evaluate their impact on the resulting quality. To fit with the augmented reality context, we propose a real-time extraction of these information from video streams, in a single pass.