Jointly Aligning Millions of Images with Deep Penalised Reconstruction Congealing

no code implementations • ICCV 2019 • Roberto Annunziata, Christos Sagonas, Jacques Cali

Extrapolating fine-grained pixel-level correspondences in a fully unsupervised manner from a large set of misaligned images can benefit several computer vision and graphics problems, e. g. co-segmentation, super-resolution, image edit propagation, structure-from-motion, and 3D reconstruction.

3D Reconstruction Super-Resolution

DeSTNet: Densely Fused Spatial Transformer Networks

no code implementations • 11 Jul 2018 • Roberto Annunziata, Christos Sagonas, Jacques Calì

In this paper, we propose Densely Fused Spatial Transformer Network (DeSTNet), which, to our best knowledge, is the first dense fusion pattern for combining multiple STNs.

Joint registration and synthesis using a probabilistic model for alignment of MRI and histological sections

no code implementations • 16 Jan 2018 • Juan Eugenio Iglesias, Marc Modat, Loic Peter, Allison Stevens, Roberto Annunziata, Tom Vercauteren, Ed Lein, Bruce Fischl, Sebastien Ourselin

Here, we overcome this limitation with a probabilistic method that simultaneously solves for registration and synthesis directly on the target images, without any training data.

Bayesian Inference