conference paper

Conference/ProceedingsPicture Coding Symposium
Start date04.12.2016
End date07.12.2016
AddressNuremberg, Germany
Author(s)Jonas Geistert, Tobias Senst, Thomas Sikora
TitleRobust Local Optical Flow: Dense Motion Vector Field Interpolation
AbstractOptical flow methods integrating sparse point correspondences have made significant contribution in the field of optical flow estimation. Especially for the goal of estimating motion accurately and efficiently sparse-to-dense interpolation schemes for feature point matches have shown outstanding performances. Concurrently, local optical flow methods have been significantly improved with respect to long-range motion estimation in environments with varying illumination. This motivates us to propose a sparse-to-dense approach based on the Robust Local Optical Flow method. We study the performance of different efficient motion vector interpolation techniques for recent optical low benchmarks. Compared state-of-the-art method the proposed approach is significantly faster while remaining competitive accuracy on Middlebury, KITTI 2015 and MPI-Sintel data-set.
Key wordsOptical Flow, RLOF,