direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Robust Local Optical Flow

The RLOFlib library is a multi-threading feature tracking library that provides access to the implementation of our recently developed PLK/RLOF based local optical flow methods. Our research is motivated by the problem of local motion estimation via robust regression with linear models. Motivated by the ongoing success of the Lucas Kanade method, we focus on optical flow based feature tracking methods with the requirements of real-time capability, accuracy and scalability.

Related Publications



  • Tobias Senst, Thilo Borgmann, Ivo Keller, Thomas Sikora
    Cross based Robust Local Optical Flow
    21th IEEE International Conference on Image Processing, Paris,France, 27.10.2014 - 30.10.2014, pp. 1967-1971
    ISBN: 978-1-4799-5750-7 DOI:10.1109/ICIP.2014.7025394
    Details BibTeX






Change Logs
stay connected via Twitter

08/01/2013 Release Version 1.1

  • Improved GPU implementation by extending image borders and enhanced the runtime performance of the CL kernels.
  • Parallelize CPU implementation using Thread Building Blocks and SSE2.

17/11/2014 Release Version 1.2

  • Add cross based and bilinear equation based variations of PLK/RLOF.
  • Remove GPU support and CUDA dependencies.
  • Support Windows, Linux and MAC OS with the following paralellization techniques

30/06/2017 Release Version 1.3

  • Add global motion prior allowing to improve the accuracy of estimation long-range motions
  • Add illumination robust model to imporve the accuracy in environments with illumination changes.
  • Intel's Threading Building Blocks (Intel TBB) is no longer supported. Parallelization will be based on SSE and OpenMP.
  • Add Python interface

... at the moment we are working on a new release including add-ons published for the newest PCS publication


The RLOFLib is now available via GitHub.

Please visit us at https://github.com/tsenst/RLOFLib

Please feel free to contact us () if you need a specific compilant for your development enviroment.

[List of people who downloaded the RLOFLib]



Terms of Use

All code is provided for research purposes only and without any warranty. Any commercial use is prohibited. By using the code in your research work, you should cite the respective paper.



Zusatzinformationen / Extras