FALDOI: Large Displacement Optical Flow by Astute Initialization

We propose a large displacement optical flow method that introduces a new strategy to compute a good local minimum of any optical flow energy functional. The method requires a given set of discrete matches, which can be extremely sparse, and an energy functional. The matches are used to guide a structured coordinate-descent of the energy functional around these keypoints. It results in a two-step minimization method at the finest scale which is very robust to the inevitable outliers of the sparse matcher, and it is better than the multi- scale methods, especially when there are small objects with very large displacements, that the multi-scale methods are incapable to find. Indeed, the proposed method recovers the correct motion field of any object which has at least one correct match, regardless of the magnitude of the displacement. We validate our proposal using several optical flow variational models. The results consistently outperform the coarse-to-fine approaches and achieve good qualitative and quantitative performance on the standard optical flow benchmarks.