Abstract. This paper proposes an original problem of stereo computation from a single mixture image– a challenging problem that had not
been researched before. The goal is to separate (i.e., unmix) a single
mixture image into two constitute image layers, such that the two layers
form a left-right stereo image pair, from which a valid disparity map can
be recovered. This is a severely illposed problem, from one input image
one effectively aims to recover three (i.e., left image, right image and
a disparity map). In this work we give a novel deep-learning based solution, by jointly solving the two subtasks of image layer separation as
well as stereo matching. Training our deep net is a simple task, as it does
not need to have disparity maps. Extensive experiments demonstrate the
efficacy of our method