On the optical flow model selection through metaheuristics.

Danillo R. Pereira*, José Delpiano, João P. Papa

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


Optical flow methods are accurate algorithms for estimating the displacement and velocity fields of objects in a wide variety of applications, being their performance dependent on the configuration of a set of parameters. Since there is a lack of research that aims to automatically tune such parameters, in this work, we have proposed an optimization-based framework for such task based on social-spider optimization, harmony search, particle swarm optimization, and Nelder-Mead algorithm. The proposed framework employed the well-known large displacement optical flow (LDOF) approach as a basis algorithm over the Middlebury and Sintel public datasets, with promising results considering the baseline proposed by the authors of LDOF.

Original languageEnglish
Article number11
JournalEurasip Journal on Image and Video Processing
Issue number1
StatePublished - 26 Dec 2015

Bibliographical note

Funding Information:
The authors are grateful to FAPESP grants #2013/20387-7 and #2014/16250-9, CNPq grants #303182/2011-3, #470571/2013-6, and #306166/2014-3, and Universidad de los Andes FAI grant #05/2013.

Publisher Copyright:
© 2015, Pereira et al.; licensee Springer.


  • Evolutionary algorithms
  • Optical flow methods
  • Optimization methods


Dive into the research topics of 'On the optical flow model selection through metaheuristics.'. Together they form a unique fingerprint.

Cite this