Robust depth estimation with occlusion detection using concepts of optical flow

Abstract

In this paper we present an approach to go beyondthe accuracy limits of current optical flow estimators. We haveused coarse to over-fine approach along with hybridinterpolation to upgrade coarse to fine approach. Coarse to fineapproach is used by most modern optical flow algorithms.Results of suggested method show benefit for sub-pixel motion.It also reduces the estimation error to great extent. Hybridinterpolation method is used which is an integration of bilinearand bi-cubic interpolation methods. Results of our approachon benchmark sequences show that estimated depth map areclearer and boundaries are sharper than original coarse tofine approach with bi-cubic interpolation method. Once opticalflow vectors are found occlusion is detected based on theconcept of residual error image.