Formation of 3D Outlines of Objects Using MOPS and SIFT Using Monocular Vision

Abstract

This paper presents how to form 3D outlines of objects in an image obtained from a monocular vision. It detects general outlines of objects to make good use of MOPS and obtains the space-coordinates of outlines. Simultaneously, it obtains the space-coordinates of feature points that exist in the internal outlines of objects through SIFT-algorithm. By merging the space-coordinates of both outlines and feature points, it generates the 3D information of objects. The method proposed in this paper considers only the approximate outline of objects, which enables rapid calculation. Also it has the advantage of being capable of collecting detailed 3D information because it uses SIFT feature points inside objects.