Dr Graham Thomas
Academic and research departments
Centre for Vision, Speech and Signal Processing (CVSSP), Surrey Institute for People-Centred Artificial Intelligence (PAI).About
Biography
I joined CVSSP as Professor of Creative Technology in April 2025, working mainly on AI4ME and CoSTAR, applying computer vision and AI to media production. I brought with me more than 40 years at BBC R&D, with my last role there being Head of Applied Research, Production. My work included leading research on computer vision & graphics for programme production and overseeing work on immersive and object-based audio, as well as running R&D’s Graduate Scheme.
I did my PhD on video motion estimation while working at the BBC back in 1990, which led to a commercially successful standards converter which won an Emmy award and a Queen’s Award and is still in use today for conversion between 50Hz and 60Hz TV standards. I went on to work on various video coding standards, then developed some of the first technology for virtual production. This included a new keying method and a camera tracking system used both for TV production and in films including Harry Potter, with the tracking system winning an RTS Technical Innovation Award. I had been collaborating with Adrian Hilton and the CVSSP team on various projects since 1999, with results contributing directly to a sports graphics system that was commercialised and has been in regular use by the BBC and many other broadcasters since 2004. The system won an IBC Production Innovation award in 2006 and a Queen’s Award for Innovation in April 2011.
I enjoy collaborating across industry and academia and have been involved in numerous EU and InnovateUK projects, including MAX-R (Horizon Europe, investigating capturing live music performances for experiencing in a game engine) and Polymersive (InnovateUK, looking at capturing audio and video for immersive media). I also helped set up the BBC Audio Research Partnership in 2011, which was the first large-scale multi-university partnership that BBC R&D was involved in.
I worked with Adrian and our partners at the University of Lancaster to set up the 5-year AI4ME Prosperity Partnership on personalised media at scale, which started in September 2021.
Areas of specialism
My qualifications
Previous roles
ResearchResearch interests
I am interested in applying advances in audio-visual capture, creation, understanding and rendering to current and future media.
Research projects
AI4ME is a 5-year Prosperity Partnership with the BBC and the University of Lancaster, researching AI technologies that transform captured audio and visual content into media objects, enabling personalised and customisable media experiences.
Research collaborations
I am working with the CoSTAR National Lab team, led by Royal Holloway, with researchers at Surrey, Abertay and the National Film and Television School, plus other CoSTAR partners. My role includes helping to connect the research work with industry.
Research interests
I am interested in applying advances in audio-visual capture, creation, understanding and rendering to current and future media.
Research projects
AI4ME is a 5-year Prosperity Partnership with the BBC and the University of Lancaster, researching AI technologies that transform captured audio and visual content into media objects, enabling personalised and customisable media experiences.
Research collaborations
I am working with the CoSTAR National Lab team, led by Royal Holloway, with researchers at Surrey, Abertay and the National Film and Television School, plus other CoSTAR partners. My role includes helping to connect the research work with industry.
Teaching
I am an examiner for PhD confirmations in the Centre for Doctoral Training in AI for Digital Media Inclusion, as well as other PhD studentships in CVSSP.
Publications
Highlights
My publications mainly cover video analysis and understanding for broadcast-related applications, such as graphics for sports analysis and 3D capture for immersive experiences. I have also been involved in work on user experience, including the impact of high dynamic range video on immersion.