Shape from symmetry

We describe a technique for reconstructing probable occluded surfaces from 3D range images. The technique exploits the fact that many objects possess shape symmetries that can be recognized even from partial 3D views. Our approach identifies probable symmetries and uses them to attend the partial 3D shape model into the occluded space. To accommodate objects consisting of multiple parts, we describe a technique for segmenting objects into parts characterized by different symmetries. Results are provided for a real-world database of 3D range images of common objects, acquired through an active stereo rig

1. Introduction

In recent years, shape completion has become a major research area in 3-D computer vision. In the shape completion problem, one is given a partial 3-D view of an object surface. The view might be acquired through a stereoscopic camera system, a laser-based range finder, or similar. The view only captures one side of the object; other surfaces remain occluded. The shape completion problem is the problem of reconstructing these occluded surfaces from the visible parts of the object.