As part of ongoing research into machine learning for imaging and computer vision at the University of Edinburgh, we are looking for a highly motivated PhD student to work on new self supervised machine learning techniques for imaging to start as soon as possible.
Computational imaging and low-level computer vision tasks, such as medical imaging or 3D vision, are extremely challenging due to an inability to directly measure all quantities of interest. Today, the state-of-the-art imaging solutions are based around machine learning and deep neural networks. However, these typically require unrealistic access to a large quantity of labelled ground truth data for training, without which such networks can exhibit severe bias and overfitting. To overcome the need for ground truth data, there has been a drive to find algorithms that can learn purely from acquired imaging data.
This project will build on recent advances in self-supervised learning developed at the University of Edinburgh and will research new theoretical and algorithmic solutions for imaging without ground truth data. The research will open up the prospect for new imaging solutions and offer the student the opportunity to work on an array of applications ranging from advanced medical and scientific imaging to the development of new self-supervised dynamic vision systems for robotics and autonomous systems.
A background understanding in one or more of the following is desirable:
- Computational Imaging
- 3D Computer vision
- Artificial intelligence/Machine Learning/Deep Learning
Informal enquiries by email can be sent to Prof. Mike Davies ([email protected] ).
Further Information:
The successful applicant will be awarded a 4 year studentship, which includes their stipend and tuition fees at the UK/EU rate, and contributions towards travel and research costs for their PhD project.
The University of Edinburgh is committed to equality of opportunity for all its staff and students, and promotes a culture of inclusivity. Please see details here: www.ed.ac.uk/equality-diversity
Closing date: Monday, January 16th , 2023