Stereo Computation for a Single Mixture Image

TL;DR

This paper introduces a novel deep learning approach to generate stereo image pairs from a single mixed image by jointly separating the layers and estimating disparity, addressing a previously unexplored problem.

Contribution

It presents the first method to perform stereo computation from a single mixture image using deep learning without requiring disparity supervision.

Findings

Effective separation of stereo layers from a single image.

Successful disparity map recovery without disparity ground truth.

Outperforms baseline methods in experiments.

Abstract

This paper proposes an original problem of \emph{stereo computation from a single mixture image}-- a challenging problem that had not been researched before. The goal is to separate (\ie, 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 (\ie, 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.