AI-Fun Seminar | Neill Campbell: Generative models as priors for inverse problems
| Dates: | 8 May 2024 |
|---|
| Times: | 10:30 - 10:30 |
|---|
| What is it: | Seminar |
|---|
| Organiser: | Faculty of Science and Engineering |
|---|
| Who is it for: | University staff, External researchers, Current University students |
|---|
| Speaker: | Neill Campbell |
|---|
|
The Manchester Centre for AI Fundamentals is hosting a series of seminars featuring expert researchers working in the fundamentals of AI.
Title: Generative models as priors for inverse problems.
Abstract: We will discuss the use of generative models in the regularisation of inverse problems involving both appearance and shape as well as the merits of integrating probabilistic approaches. The different natures of appearance and shape motivate the need for fundamentally different modelling approaches and we will contrast effective models as well as consider how to unify them towards a universal modelling framework that generalises to a range of desirable properties. We will illustrate the utility of this methodology across a range of real-world problems from medical imaging to the creative industries.
Bio: Neill Campbell is a Royal Society Industry Fellow and Professor of Visual Computing and Machine Learning in the Department of Computer Science at the University of Bath. He is the director of the Centre for the Analysis of Motion, Entertainment Research and Applications (CAMERA) that researches and applies visual computing and machine learning technology in the fields of Entertainment, Health and Sports Science; chair of the British Machine Vision Association; and co-director of the Centre for Mathematics and Algorithms for Data, an inter-disciplinary group that studies the theoretical underpinnings of Machine Learning and Data Science; and the director of research for MyWorld, a creative hub across the Bath and Bristol region, funded by UKRI and an alliance of more than 30 industry and academic partners. Before moving to Bath he worked as a post-doc at UCL and Cambridge, where he also completed his PhD in Computer Vision. His research involves learning models of shape, appearance and dynamics from images and video. In particular, creating systems that do not require technical computing expertise (e.g. for artists). He also works on machine learning problems where data are scarce or expensive to obtain (e.g. annotations from expert clinicians) and when uncertainty in the resulting output is important (e.g. medical and safety applications).
Travel and Contact Information
Find event
Emmeline Suite
Christabel Pankhurst Building
Dover Street
Manchester
