Visual Hull based 3D Reconstruction for Virtual World

In this project, we design an algorithm to reconstruct a 3D skeleton from several images of the object.

The visual hull is the intersection of the visual cones formed by back-projecting the silhouettes found in the corresponding images from a set of cameras. An image-based algorithm for computing visual hulls using oriented projective constraints is adopted. The 3D reconstruction is extended to virtual world using simple collision laws.