PROJECT TITLE :

Continuous Depth Map Reconstruction From Light Fields - 2015

ABSTRACT:

Lightweight field analysis recently received growing interest, since its wealthy structure information edges several laptop vision tasks. This paper presents a completely unique methodology to reconstruct continuous depth maps from light-weight field data. Conventional approaches sometimes treat depth map reconstruction as an optimization problem with discrete labels. On the contrary, our proposed method will acquire continuous depth maps by solving a linear system, which preserves richer details compared with typical discrete approaches. Structure tensor is utilized to extract raw depth information and corresponding confidence levels from the light field knowledge. We introduce a methodology to cut back the adverse impact of unreliable native estimations, which helps to induce rid of errors in specular areas and edges where depth values are discontinuous. Experiments on both artificial and real lightweight field information demonstrate the effectiveness of the proposed methodology.