Applications are invited for a PhD (via MPhil) studentship in the Computer Vision and Machine Learning (CVML) Research Group within the School of Engineering. The studentship is funded through the UCLan Doctoral Training Centre for Industry Collaboration and is tenable for up to 3.5 years full time [subject to satisfactory progress].  Both Home and EU/International Applicants may apply but EU/International Applicants will be required to pay the difference in tuition fees between the UK and EU/International fee rates. The studentship will provide successful applicants with an annual stipend in line with UKRI rates (Currently £17,668 per year), subject to satisfactory progress.

Project Description:

The aim of this project is to research and develop methods for automatic analysis of images and videos with objectives of detection, segmentation, and categorisation of objects present in the observed scene. The important goal of the project will be to reconstruct 3D information and provide natural language description of the observed scene. The project will develop new computational methods and their software implementation using computer vision and machine learning methodologies.

The project will build on the results from a recently completed STFC CDN+ funded AIdDeCo project. At the MPhil phase, the objective will be to test methods developed as part of the AIdDeCo (e.g., FCBformer software available on GitHub) at scale on heterogeneous data, therefore progressing it closer to clinical evaluation.

The PhD phase of the project will expand on the results from AIdDeCo. The project will develop innovative visual aids supporting colon examination. These will include 3D structure visualisation as well as navigation tools based on visual odometry techniques. The project will achieve these results by leveraging computational approaches including machine learning methods. For example, for the 3D reconstruction it will build on a non-rigid structure from motion approach, previously developed at UCLan’s Computer Vision and Machine Learning (CVML) group, where a 3D deformable scene is reconstructed from the moving monoscopic camera and implicitly learned scene deformation model. These, together with the automatic polyp detection, segmentation, and categorisation, should be instrumental in reducing risks of polyps being missed during screening.

Further information

Informal project related enquiries about the post can be made to Prof Bogdan Matuszewski via email: bmatuszewski1@uclan.ac.uk

For the full details go to: https://www.uclan.ac.uk/research/research-degrees/studentships

Applications should be completed on our online application system, selecting the following options:

• Type of Study – Research Degree (Postgraduate)
• Course - Master of Philosophy/Doctor of Philosophy
• Campus – Preston Campus
• Mode of Study – Full Time

Please quote the Studentship reference number DTC12-22-38 on the online application form in the personal statement section. You will not be able to enter the number in the Studentship reference section.

Closing Date: 31st March 2023

Proposed Interview Date: TBC

Expected Start Date: September 2023

CURRENT UCLAN RESEARCH STUDENTS WILL NOT BE ELIGIBLE TO APPLY FOR THIS RESEARCH STUDENTSHIP