Skip to Main content Skip to Navigation
Conference papers

Self-similar regularization of optic-flow for turbulent motion estimation

Patrick Héas 1, 2 Etienne Mémin 1 Dominique Heitz 2
1 VISTA - Vision spatio-temporelle et active
IRISA - Institut de Recherche en Informatique et Systèmes Aléatoires, Inria Rennes – Bretagne Atlantique
Abstract : Based on self-similar models of turbulence, we propose in this paper a multi-scale regularizer in order to provide a closure to the optic-flow estimation problem. Regularization is achieved by constraining motion increments to behave as a self-similar process. The associate constrained minimization problem results in a collection of first-order optic-flow regularizers acting at the different scales. The problem is optimally solved by taking advantage of lagrangian duality. Furthermore, an advantage of using a dual formulation, is that we also infer the regularization parameters. Since, the self-similar model parameters observed in real cases can deviate from theory, we propose to add in the algorithm a bayesian learning stage. The performance of the resulting optic-flow estimator is evaluated on a particle image sequence of a simulated turbulent flow. The self-similar regularizer is also assessed on a meteorological image sequence.
Mots-clés : REGULARISATION
Document type :
Conference papers
Complete list of metadata

Cited literature [15 references]  Display  Hide  Download

https://hal.inria.fr/inria-00325807
Contributor : Peter Sturm <>
Submitted on : Tuesday, September 30, 2008 - 1:46:42 PM
Last modification on : Thursday, January 7, 2021 - 4:21:09 PM
Long-term archiving on: : Monday, October 8, 2012 - 1:45:21 PM

File

mlvma08_submission_9.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00325807, version 1

Citation

Patrick Héas, Etienne Mémin, Dominique Heitz. Self-similar regularization of optic-flow for turbulent motion estimation. The 1st International Workshop on Machine Learning for Vision-based Motion Analysis - MLVMA'08, Oct 2008, Marseille, France. ⟨inria-00325807⟩

Share

Metrics

Record views

442

Files downloads

627