Toward non-intrusive motion capture

Shape-from-silhouettes, a simple approach to 3D shape understanding, can also be used for recovering posture and motion of human bodies. Silhouettes are easy to obtain from intensity images, do not require foreign objects attached to the subject, can directly provide a 3-D reconstruction of the body, or drive model-based motion capture. The purpose of this work is to demonstrate the effectiveness of this approach in a virtual environment, and to investigate number and position of stationary cameras suitable for precise reconstruction of the 3-D posture and motion of the human body.