The project: Applications are invited for a fully-funded PhD studentship in Computer Vision and Machine Learning. The successful candidate will be working on a project focused on zero-shot activity recognition. During the PhD, the student will learn and use computer vision methods such as tracking, object detection and localisation, action and higher-level activity recognition. Machine learning methods related to transfer learning and Long Short-Term Memory (LSTM) neural networks will be used as well as conventional convolutional neural networks (CNN). The student will also learn how to analyse and work with images captured using RGB-D sensors (e.g. Kinect V2).

Problem Statement: A single visual and depth sensor, deployed in e.g. a kitchen, will observe a person performing their daily chores, such as washing the dishes, preparing a meal or cleaning. By learning these activities from one or more different kitchens, where people perform their chores differently, this project will target automatically identifying the expected activities in the new kitchen. As more (unlabelled) data is gathered, the solution aims to perform with accuracy comparable to that trained on labelled data in the new kitchen itself.

Applicant will work in a vibrant computer vision lab, with more than 15 PhD students and 5 postdoctoral researchers working in closely related topics. For an insight into the supervisor’s current and previous works, refer to: http://www.cs.bris.ac.uk/~damen

How to apply: Applicants should first send their CV (max 4 pages) to Dr. Damen ([email protected]) before they are considered for this studentship. If they meet the minimum criteria, they will be asked to submit a full application with academic references online at:http://www.bristol.ac.uk/study/postgraduate/apply/.

Please ensure that in the Funding section you tick “I would like to be considered for a funding award from the Computer Science Department” and specify the title of the scholarship in the “other” box below with the name of the supervisor.

Candidate requirements:  A minimum 2.1 honours degree or equivalent in Computer Science, Mathematics or other relevant subjects. Standard English language requirement: Profile E

Basic skills and knowledge required include solid mathematical ability with excellent programming skills; Machine Learning and Computer Vision

Funding: This EPSRC scholarship will cover full UK/EU (EU applicants who have been resident in the UK for 3 years prior to application) PhD tuition fees and a tax-free stipend at the current RCUK rate (£14,296 in 2016/17). EU nationals resident in the EU may also apply and though they will qualify only for PhD tuition fees, additional funding will be matched from an alternative source

Contacts: For informal enquiries, contact Dr. Dima Damen: [email protected]

For general enquiries, please email [email protected]