3 papers
cs.CV2026
Dual-domain U-Nets with embedded back projection operators for motion-resolved 4D CBCT reconstruction
Ivo Herzig, Pascal Paysan, Daniel Barco +10
Four-dimensional cone beam CT (4D CBCT) is important for image-guided radiation therapy of thoracic cancers, but its use is limited by long scan times, causing high patient dose an…
eess.IV2026
Projected Energy Matching for Generative 3D Priors
Daniel Barco, Michal Balcerak, Suprosanna Shit +4
Energy Matching has emerged as a powerful generative framework that combines flow model efficiency with the explicit likelihood of Energy-Based Models (EBMs) via a single, time-ind…
cs.CV2025
MInDI-3D: Iterative Deep Learning in 3D for Sparse-view Cone Beam Computed Tomography
Daniel Barco, Marc Stadelmann, Martin Oswald +7
We present MInDI-3D (Medical Inversion by Direct Iteration in 3D), the first 3D conditional diffusion-based model for real-world sparse-view Cone Beam Computed Tomography (CBCT) ar…