Detection of Curvilinear Solid Object with Three Dimensional Voting

Abstract

In this paper, a method to detect position and orientation of an object is proposed; it matches three-dimensionally an object to its model object by repeating vote for the exact position and orientation of the object using three arbitrary points chosen from input data. Since both template data and input data are only information related a 3-dimensional point, it is simple for creation and acquisition of data, and has fewer influences from noise.