Bayesian optical flow with uncertainty quantification
From MaRDI portal
Abstract: Optical flow refers to the visual motion observed between two consecutive images. Since the degree of freedom is typically much larger than the constraints imposed by the image observations, the straightforward formulation of optical flow as an inverse problem is ill-posed. Standard approaches to determine optical flow rely on formulating and solving an optimization problem that contains both a data fidelity term and a regularization term, the latter effectively resolves the otherwise ill-posedness of the inverse problem. In this work, we depart from the deterministic formalism, and instead treat optical flow as a statistical inverse problem. We discuss how a classical optical flow solution can be interpreted as a point estimate in this more general framework. The statistical approach, whose "solution" is a distribution of flow fields, which we refer to as Bayesian optical flow, allows not only "point" estimates (e.g., the computation of average flow field), but also statistical estimates (e.g., quantification of uncertainty) that are beyond any standard method for optical flow. As application, we benchmark Bayesian optical flow together with uncertainty quantification using several types of prescribed ground-truth flow fields and images.
Recommendations
Cites work
- scientific article; zbMATH DE number 3551792 (Why is no real title available?)
- scientific article; zbMATH DE number 194343 (Why is no real title available?)
- scientific article; zbMATH DE number 3291403 (Why is no real title available?)
- A framework for estimating potential fluid flow from digital imagery
- A lagged diffusivity method for computing total variation regularized fluid flow
- A multi-time step method to compute optical flow with scientific priors for observations of a fluidic system
- A nonlocal Bayesian image denoising algorithm
- A proof of convergence of the Horn-Schunck optical flow algorithm in arbitrary dimension
- A statistical perspective on ill-posed inverse problems (with discussion)
- Bayesian data analysis.
- Computational Methods for Inverse Problems
- Computer vision. Algorithms and applications
- Computing Optical Flow via Variational Techniques
- Determining optical flow
- Gaussian Markov random field priors for inverse problems
- Large scale Bayesian inference and experimental design for sparse linear models
- MCMC-based image reconstruction with uncertainty quantification
- Nonparametric statistical inverse problems
- Parallel iterative methods for sparse linear systems
- Solving ill-posed image processing problems using data assimilation
- Space and space-time modeling using process convolutions
- Statistical and computational inverse problems.
- Tolerance regions for a multivariate normal population
- Variational optic flow computation with a spatio-temporal smoothness constraint
- Variational-Bayes optical flow
Cited in
(6)- Learning the temporal evolution of multivariate densities via normalizing flows
- A unified probabilistic model of the perception of three-dimensional structure from optic flow
- Chilled sampling for uncertainty quantification: a motivation from a meteorological inverse problem *
- Parametric approach to promote a divergence-free flow in the image-based motion estimation with application to bioirrigation
- Variational-Bayes optical flow
- UNLOC: optimal unfolding localization from noisy distance data
This page was built for publication: Bayesian optical flow with uncertainty quantification
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4582729)
