Bayesian estimation of turbulent motion

Based on physical laws describing the multiscale structure of turbulent flows, this paper proposes a regularizer for fluid motion estimation from an image sequence. Regularization is achieved by imposing some scale invariance property between histograms of motion increments computed at different sca...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Heás, P., Herzet, C., Meḿin, E., Heitz, D., Mininni, P.D.
Formato: JOUR
Materias:
Bayesian model selection
Constrained optimization
Optic flow
Robust estimation
Turbulence
Bayesian perspective
Fluid flow estimation
Fluid motion estimation
Multi-scale structures
Scale-invariance property
Bayesian networks
Computer vision
Estimation
Flow of fluids
Models
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_01628828_v35_n6_p1343_Heas
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
id todo:paper_01628828_v35_n6_p1343_Heas
record_format dspace
spelling todo:paper_01628828_v35_n6_p1343_Heas2023-10-03T15:01:36Z Bayesian estimation of turbulent motion Heás, P. Herzet, C. Meḿin, E. Heitz, D. Mininni, P.D. Bayesian model selection Constrained optimization Optic flow Robust estimation Turbulence Bayesian model selection Bayesian perspective Fluid flow estimation Fluid motion estimation Multi-scale structures Optic flow Robust estimation Scale-invariance property Bayesian networks Computer vision Constrained optimization Estimation Flow of fluids Turbulence Models Based on physical laws describing the multiscale structure of turbulent flows, this paper proposes a regularizer for fluid motion estimation from an image sequence. Regularization is achieved by imposing some scale invariance property between histograms of motion increments computed at different scales. By reformulating this problem from a Bayesian perspective, an algorithm is proposed to jointly estimate motion, regularization hyperparameters, and to select the most likely physical prior among a set of models. Hyperparameter and model inference are conducted by posterior maximization, obtained by marginalizing out non-Gaussian motion variables. The Bayesian estimator is assessed on several image sequences depicting synthetic and real turbulent fluid flows. Results obtained with the proposed approach exceed the state-of-the-art results in fluid flow estimation. © 2013 IEEE. Fil:Mininni, P.D. Universidad de Buenos Aires. Facultad de Ciencias Exactas y Naturales; Argentina. JOUR info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by/2.5/ar http://hdl.handle.net/20.500.12110/paper_01628828_v35_n6_p1343_Heas
institution Universidad de Buenos Aires
institution_str I-28
repository_str R-134
collection Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA)
topic Bayesian model selection
Constrained optimization
Optic flow
Robust estimation
Turbulence
Bayesian model selection
Bayesian perspective
Fluid flow estimation
Fluid motion estimation
Multi-scale structures
Optic flow
Robust estimation
Scale-invariance property
Bayesian networks
Computer vision
Constrained optimization
Estimation
Flow of fluids
Turbulence
Models
spellingShingle Bayesian model selection
Constrained optimization
Optic flow
Robust estimation
Turbulence
Bayesian model selection
Bayesian perspective
Fluid flow estimation
Fluid motion estimation
Multi-scale structures
Optic flow
Robust estimation
Scale-invariance property
Bayesian networks
Computer vision
Constrained optimization
Estimation
Flow of fluids
Turbulence
Models
Heás, P.
Herzet, C.
Meḿin, E.
Heitz, D.
Mininni, P.D.
Bayesian estimation of turbulent motion
topic_facet Bayesian model selection
Constrained optimization
Optic flow
Robust estimation
Turbulence
Bayesian model selection
Bayesian perspective
Fluid flow estimation
Fluid motion estimation
Multi-scale structures
Optic flow
Robust estimation
Scale-invariance property
Bayesian networks
Computer vision
Constrained optimization
Estimation
Flow of fluids
Turbulence
Models
description Based on physical laws describing the multiscale structure of turbulent flows, this paper proposes a regularizer for fluid motion estimation from an image sequence. Regularization is achieved by imposing some scale invariance property between histograms of motion increments computed at different scales. By reformulating this problem from a Bayesian perspective, an algorithm is proposed to jointly estimate motion, regularization hyperparameters, and to select the most likely physical prior among a set of models. Hyperparameter and model inference are conducted by posterior maximization, obtained by marginalizing out non-Gaussian motion variables. The Bayesian estimator is assessed on several image sequences depicting synthetic and real turbulent fluid flows. Results obtained with the proposed approach exceed the state-of-the-art results in fluid flow estimation. © 2013 IEEE.
format JOUR
author Heás, P.
Herzet, C.
Meḿin, E.
Heitz, D.
Mininni, P.D.
author_facet Heás, P.
Herzet, C.
Meḿin, E.
Heitz, D.
Mininni, P.D.
author_sort Heás, P.
title Bayesian estimation of turbulent motion
title_short Bayesian estimation of turbulent motion
title_full Bayesian estimation of turbulent motion
title_fullStr Bayesian estimation of turbulent motion
title_full_unstemmed Bayesian estimation of turbulent motion
title_sort bayesian estimation of turbulent motion
url http://hdl.handle.net/20.500.12110/paper_01628828_v35_n6_p1343_Heas
work_keys_str_mv AT heasp bayesianestimationofturbulentmotion
AT herzetc bayesianestimationofturbulentmotion
AT memine bayesianestimationofturbulentmotion
AT heitzd bayesianestimationofturbulentmotion
AT mininnipd bayesianestimationofturbulentmotion
_version_ 1807314654869848064

Ejemplares similares