ESTIMATION OF CAMERA 3D-POSITION TO MINIMIZE OCCLUSIONS

Abstract

Occlusions are almost always seen as undesirable singularities that pose difficult challenges to recognition\n\t\t\t\t processes of objects which have to be manipulated by a robot. Often, the occlusions are perceived because\n\t\t\t\t the viewpoint with which a scene is observed is not adapted. In this paper, a strategy to determine the\n\t\t\t\t location, orientation and position, more suitable so that a camera has the best viewpoint to capture a scene\n\t\t\t\t composed by several objects is presented. The estimation for the best location of the camera is based on\n\t\t\t\t minimizing the zones of occlusion by the analysis of a virtual image sequence in which is represented the\n\t\t\t\t virtual projection of the objects. These virtual projections represent the images as if they were captured by a\n\t\t\t\t camera with different viewpoints without moving it.