Introduction - If you have any usage issues, please Google them yourself
The code implements a coarse-to-fine variational framework for optical flow estimation between two image frames that incorporates descriptor matches in additiona to the standard brightness and gradient constrancy constraints, for dealing effectively with large displacements of small (and large) structures. The code is based on [1] and [2]. We denote with w(x)=(u,v) the displacement field between two images I_1 and I_2. Following [2], we obtain descriptor matches w_descr(x) by matching densely sampled HOG descriptors in the two images via approximate nearest neighbor search. We then compute w(x) by minimizing the energy functional: