Real time on board stereo camera pose through image registration

Abstract

This work addresses the real time estimation of on-board stereo headpsilas position and orientation. Unlike existing works which rely on feature extraction either in the image domain or in 3D space, our approach directly estimates the unknown parameters from the stream of stereo pairspsila brightness. The pose parameters are estimated by registering the right and left images, which is solved using two optimization techniques: the differential evolution algorithm and the Levenberg-Marquardt algorithm. We provide experiments and evaluations of performance.