Vision, Speech and Signal Processing PhD

Why choose this course

At the Centre for Vision, Speech and Signal Processing (CVSSP) we’re developing exciting and ground-breaking technologies, from facial recognition for security and medical imaging through to 3D spatial audio and 3D reconstruction from video for visual-effects production in films, games and virtual reality. Our research is creating machines that can see, hear and understand the world around them.

Our research aims to advance the state-of-the-art in audio-visual signal processing, computer vision and machine learning, with a focus on image, video and audio applications. We have expertise and activities in computer vision, digital signal processing, machine learning and artificial intelligence, computer graphics and human computer interaction, data science for media, medical image analysis and multimedia communication.

We often collaborate with industry and our research has led to frequent technology transfer and exploitation, with previous research projects having resulted in award-winning spin-out companies in biometrics, communications, medical technologies and in the creative industries.

Our audio-visual research facilities are cutting-edge, and include a unique multiple UltraHD performance capture studio and high-performance computing capabilities for real-time analysis and machine learning.

What you will study

You can take a maximum of four years of full-time study to complete our PhD programme. After 12 months, you’ll write a confirmation report, which is assessed by two independent examiners. After that, you’ll submit a written PhD thesis after a minimum of three years of full-time study. We’ll work with you to make a detailed evaluation of your progress and future research plans every six months to help you keep on track to successfully complete your PhD.

You’ll be allocated two Surrey-based academic supervisors, in addition to any external collaborative supervisors. Your principal supervisor will be an expert in your area of research and will be in charge of monitoring your research progress on a regular basis. Your supervisors will help you to define the initial objective and scope of your research, and to refine these as your project evolves. They’ll direct you to resources to help you learn the necessary experimental, theoretical or computing skills you need, and will be able to advise you on how to complete your PhD and your thesis. We often appoint external collaborative supervisors to contribute specific expertise or to allow access to external resources or organisations.

Research themes

Our research activities within the CVSSP are grouped into the following labs:

• The A-Lab focuses on machine audition, developing audio signal processing technology relating to speech recognition, and blind source separation.
• The B-Lab focuses on biometric and security related technologies, specialising in facial biometrics and lip tracking.
• The C-Lab focuses on cognitive systems, covering a broad range of technologies relating to visual interaction with computers. These include sign language recognition and robotics.
• The I-Lab focuses on multimedia communications, including audio-visual coding, annotation, search, broadcast and streaming technologies.
• The M-Lab focuses on medical imaging technologies and works closely with leading healthcare institutions.