R
Riccardo Barbano
Researcher at University College London
Publications - 20
Citations - 77
Riccardo Barbano is an academic researcher from University College London. The author has contributed to research in topics: Computer science & Iterative reconstruction. The author has an hindex of 2, co-authored 10 publications receiving 16 citations.
Papers
More filters
A Probabilistic Deep Image Prior over Image Space
TL;DR: The deep image prior regularises under-specified image reconstruction problems by reparametrising the target image as the output of a CNN using a classical image reconstruction regulariser and an efficient linearised Laplace inference algorithm is proposed.
Proceedings ArticleDOI
Quantifying Model Uncertainty in Inverse Problems via Bayesian Deep Gradient Descent
TL;DR: In this article, a scalable, data-driven, knowledge-aided computational framework was developed to quantify the model uncertainty via Bayesian neural networks, where only the last layer of each block is Bayesian, while the others remain deterministic.
Posted Content
Quantifying Sources of Uncertainty in Deep Learning-Based Image Reconstruction
TL;DR: This work proposes a scalable and efficient framework to simultaneously quantify aleatoric and epistemic uncertainties in learned iterative image reconstruction, which builds on a Bayesian deep gradient descent method for quantifying epistemic uncertainty, and incorporates the heteroscedastic variance of the noise to account for the aleATORic uncertainty.
Journal ArticleDOI
Unsupervised knowledge-transfer for learned image reconstruction*
TL;DR: In this article , an unsupervised knowledge-transfer paradigm for learned reconstruction within a Bayesian framework is proposed, where the first phase trains a reconstruction network with a set of ordered pairs comprising of ground truth images of ellipses and the corresponding simulated measurement data.
Uncertainty Estimation for Computed Tomography with a Linearised Deep Image Prior
Javier Antor'an,Riccardo Barbano,Johannes Leuschner,Jos'e Miguel Hern'andez-Lobato,Bangti Jin +4 more
TL;DR: In this paper , the authors propose the linearised deep image prior (DIP) to estimate the uncertainty associated with reconstructions produced by the DIP with total variation regularisation (TV).