The KITTI Vision Benchmark Suite

Most variational optical flow methods are based on the well-known brightness constancy assumption or high-order constancy assumptions in order to implement the data term in the optimization energy function. Unfortunately, any variation in the lighting within the scene violates the brightness constancy constraint; in turn, the gradient constancy assumption does not work properly with large illumination changes. This paper proposes an illumination-robust constancy based on a robust texture descriptor rather than the brightness constancy. Thus, the similarity function used as a data term was obtained from extracting texture features through the local directional pattern descriptor for two consecutive frames within the duality total variational optical flow algorithm. In addition, a weighted non-local term that depends on both the color similarity and the occlusion state of pixels is integrated during the optimization process in order to increase the accuracy of the resulting flow field. Experimental results show a qualitative comparison with the proposed approach and yield state-of-the-art results on the KITTIDIFaddbegin DIFadd{, Midleburry and MPI-sintel }DIFaddend datasets.

\lambda = 10;
Pyramid_Scale = 0.9;
Warp/Level = 4;
Fixed_point_iteration/Warp = 10;

@article { LDPOF,
author = { Mahmoud A. Mohamed and Hatem A. Rashwan and B{"a}rbel Mertsching and Miguel Angel Garcia and Domenec Puig },
title = { Illumination-Robust Optical Flow Using Local Directional Pattern },
month = { February },
year = { 2014 },
journal = { IEEE Transactions on Circuits and Systems for Video Technology }
}