Reflection Separation using a Pair of Unpolarized and Polarized Images

Youwei Lyu, Zhaopeng Cui, Si Li, Marc Pollefeys, Boxin Shi

Abstract: When taking photos through glass windows or doors, the transmitted background scene is often blended with undesirable reflection. Separating two layers apart to enhance the image quality are of vital importance for both human and machine perception. In this paper, we propose to exploit physics constraints from a pair of unpolarized and polarized images to separate reflection and transmission layers. Due to the simplified capturing setup, the system becomes more underdetermined comparing with existing polarization based solution taking three or more images as input. We propose to solve semireflector orientation estimation first to make the physical image formation well-posed first and then learn to reliably separate two layers using a refinement network with gradient loss. Quantitative and qualitative experiment results show our approach performs favorably over existing polarization and single image based solutions.

Code Link: https://github.com/YouweiLyu/reflection_separation_with_un-polarized_images

CMT Num: 8242