Publication,
Research,
Docs and Info,
KAMEDA
Since our objective is to reconstruct a certain real space where people work, it is reasonable to have a knowledge of static objects in the space in advance. As the static objects do not change their positions and shapes, we can exclude the subspace where the static objects occupy. We call the subspace as static object occupation subspace (SOOS) denoted by . If we see the target space from the viewpoint of camera i, some subspaces cannot be seen because occludes them. We merge these occluded subspaces into and call it static object influence subspace (SOIS) .
From now on, we concentrate on reconstructing the voxels which represent dynamic objects in the target space.