Machine Learning

In addition to the projects highlighted below, we are part of the project Maths4DL - Mathematics for Deep Learning. It aims to put deep learning onto a firm mathematical grounding, and will combine theory, modelling, data and computation to help unlock the next generation of deep learning. This project is in collaboration with the University of Bath and University College London.

Image Classification Under Minimal Supervision: Graph-Based Semi-Supervised Learning for Real-World Large-Scale Problems.

Reseacher: Angelica Aviles-Rivero, and Carola-Bibiane Schönlieb

Learning Sampling Patterns for MRI

Researcher: Ferdia Sherry, Erlend Riis, Luca Calatroni, and Carola-Bibiane Schönlieb

Machine learning methods for segmentation of food microscopy images

Researcher: Rihuan Ke, Carola-Bibiane Schönlieb and Peter Schuetz

Model-Based and Deep Learning-Based Approaches for Image Restoration for Real-World Problems: Image Reflection Removal and Image Deraining.

Researcher: Angelica Aviles-Rivero, and Carola-Bibiane Schönlieb

Semi-Supervised Hyperspectral Image Classification

Researcher: Angelica Aviles Rivero, Philip Sellars, David Coomes, Nicolas Papadakis, Carola-Bibiane Schönlieb

Unsupervised Learning for Image Analysis

Researcher: Angelica Aviles-Rivero, Hankui Peng, Lihao Liu, and Carola-Bibiane Schönlieb

Conditional Image Generation with Score-Based Diffusion Models

Researcher: Georgios Batzolis, Jan Stanczuk, Carola-Bibiane Schonlieb and Christian Etmann.

Dynamical Systems-based Approaches to Designing Robust Neural Networks

Researcher: Ferdia Sherry

Cambridge Image Analysis