1

his paper presents a novel interactive approach for line art colorization using conditional Diffusion Probabilistic Models (DPMs). In our proposed approach, the user provides initial color strokes for colorizing the line art….

This article presents a deep learning approach for exemplar-based image colorization. It propose a novel attention block called super-attention, which is computed from superpixel features….

We proposes a new method for efficiently matching high-resolution feature maps from CNNs using attention mechanisms, which relies on a superpixel-based pooling dimensionality reduction strategy….

This paper addresses the problem of reconstructing an image from low-count Positron Emission Tomography (PET) data. We build on recent advances combining deep neural networks with expectation-maximization algorithms.