Abstract
As a key component in many computer vision system-s, optical flow estimation, especially with large displace-ments, remains an open problem. In this paper we present a simple but powerful matching method works in a coarse-to-fine scheme for optical flow estimation. Inspired by thenearest neighbor field (NNF) algorithms, our approach,called CPM (Coarse-to-fine PatchMatch), blends an effi-cient random search strategy with the coarse-to-fine schemefor optical flow problem. Unlike existing NNF techniques,which is efficient but the results is often too noisy for opticflow caused by the lack of global regularization, we propose a propagation step with constrained random search radius between adjacent levels on the hierarchical architecture. The resulting correspondences enjoys a built-in smoothing effect, which is more suited for optical flow estimation than NNF techniques. Furthermore, our approach can also cap-ture the tiny structures with large motions which is a problem for traditional coarse-to-fine optical flow algorithms. Interpolated by an edge-preserving interpolation method (EpicFlow), our method outperforms the state of the art on MPI-Sintel and KITTI, and runs much faster than the com-peting methods.