Generalized fusion moves for continuous label optimization

摘要

Energy-minimization methods are ubiquitous in computer vision and related fields. Low-level computer vision problems are typically defined on regular pixel lattices and seek to assign discrete or continuous values (or both) to each pixel such that a combined data term and a spatial smoothness prior are minimized. In this work we propose to minimize difficult, non-convex energies over continuous unknowns by repeated generalized fusion moves. In contrast to standard fusion moves, the fusion step jointly optimizes over binary and continuous sets of variables representing label ranges. Further, each fusion step can optimize over additional continuous unknowns appearing in the energy. We demonstrate the general method on a variational-inspired stereo approach, and optionally optimize over radiometric changes between the images as well.