EXPLOITING FEATURE MATCHING IN 3D RECONSTRUCTION FROM MULTIPLE IMAGES USING GVC

Abstract

The improved GVC Algorithm (Yadav, K. et al., 2012) proposed aims at high quality of 2-D to 3-D reconstruction by incorporating additional check of edge detection and usage of Histogram- based consistency as a photo consistency measure. In this paper, we focus on features of a scene from (Yadav K. et al., 2012) methodology. The reconstruction of a complex scene from multiple images is a fundamental problem in the field of computer vision. Volumetric methods of scene reconstruction are considered, and GVC is used as the base algorithm. Whenever any 2-D image is used to reconstruct a volume based object (voxels), the features in the images make color consistency check to fail i.e. carving of some required voxels. This inability to detect the color variance along the edges is faced by majority of the models proposed prior. Our approach comprises of feature matching and volumetric method GVC to obtain a good quality reconstruction.