MSc (by research) studentship: Statistical 3D Shape Modelling of the Human Head (in partnership with Alder Hey Hospital)

The aim is to use a collection of 3D images of the human head to build a statistical models of human head (and ear/eye) shape and its variation through a population. Specifically, the type of models that we are interested in are called 3D morphable models (3DMMs), which consist of a mean shape and a set of principal modes of variation constructed from Principal Component Analysis (PCA). PCA requires alignment and normalisation of the dataset of 3D images and these processes can be made more effective by application of modern Machine Learning techniques, such as deep learning, in particular convolutional neural networks. Once built, the 3DMMs find applications in Biometric recognition, medical image analysis and computer graphics (avatars).

Within the MSc, we aim to put the constructed 3DMM online so that it can be queried in a graphical interface. For example, when visiting a web page, one could click a pair of points on a human head model and the web page would then present population statistics on that particular measurement. Additionally, we aim to provide interfaces to the models so that they can be used by Craniofacial Clinicians, in particular we are collaborating with Christian Duncan at Alder Hey Children’s Hospital, Liverpool and we have linked in with their Headspace Project. The project is sponsored by a Google Faculty Award.

Research Supervision
If successful, you will conduct your research under the supervision of Dr Nick Pears (https://www-users.cs.york.ac.uk/~nep/), and you are likely to interact with other academics working in this area including Dr Will Smith and Dr Suresh Manandhar, in addition to several PhD students. Christian Duncan (https://www.researchgate.net/profile/Christian_Duncan), head of the Craniofacial Unit at Alder Hey Hospital, Liverpool, will provide guidance on medical applications of the shape models.


Applicants
This MSc by Research is suitable for Computer Science (or related) graduates who are interested in the fields of Computer Vision and Machine Learning. Candidates should have a good first degree at Bachelors or Integrated Masters level in a suitable subject area, preferably Computer Science, but others will be considered, such as Electronics or Mathematics. Overseas applicants may apply but should have additional funding in place to top up to the overseas fee level (see below).

Award funding
If successful, you will be supported for 12 months. Funding includes:

• £14,553 (2017/18 rate) per year stipend
• Home/EU tuition fees


Apply for this studentship
1. Apply to study
• You must apply online for a full-time MSc in Computer Science (by research) - https://www.york.ac.uk/study/postgraduate/courses/apply?course=DRMCOMSSCI1
• You must quote the project title (Headspace Online Studentship) in your application.
• There is no need to write a full formal research proposal (2,000-3,000 words) in your application to study, as this studentship is for a specific project.

2. Provide a personal statement
• As part of your application please provide a personal statement of 500-1,000 words with your initial thoughts on the research topic.

Deadlines
Application will remain open until a suitable candidate has been appointed.
Interviews are expected to take place within approx. 14 days of application.
The studentship will begin in 2018.

Informal enquiries
Project enquiries
Dr Nick Pears
[Email Address Removed]
+44 (0)1904 325658)

Application enquiries
[Email Address Removed]
+44 (0)1904 325404