A Study on 3D Reconstruction by Fuzzy LMedS Algorithm

Abstract

The computer vision involves many modeling problems with preventing noise caused by sensing units such as cameras. In order to improve computer vision system performance, a robust modeling technique must be developed for essential models in the system. The RANSAC and least median of squares (LMedS) algorithm have been widely applied for such issues. However, the performance deteriorates as the noise ratio increases and the modeling time for algorithms tends to increase in industrial applications. As an effective technique, we proposed a new fuzzy LMedS method based on reinforcement learning concept for robust modeling. In this study, we investigate an application of the fuzzy LMedS method to 3D shape reconstruction problem. Through numerical experiments using 3D measurement data, the performance was evaluated. Their results found the proposed method to be promising for 3D reconstruction.