Skip to main content

CfACTs Research Fellow (Framestore 1 Project) - Application of Machine Learning to Solve VFX Light Transport Situations (Fixed Term)

Key information

Salary: Salary £50,317.00 per annum (pro rata)

Date advertised: Monday, June 7, 2021

Closing date: Sunday, August 22, 2021 - 23:30

Please quote reference: FMC232

Bournemouth University’s vision is worldwide recognition as a leading university for inspiring learning, advancing knowledge and enriching society through the fusion of education, research and practice. Our highly skilled and creative workforce is comprised of individuals drawn from a broad cross section of the globe, who reflect a variety of backgrounds, talents, perspectives and experiences that help to build our global learning community.

The Faculty of Media and Communication at Bournemouth University is one of the largest of its kind in the world and has a global reputation for combining research excellence and teaching practice.  The research fellow will join the Faculty’s Centre for Applied Creative Technologies (CfACTs), a Marie Skłodowska-Curie Actions (MSCA) COFUND research training centre that provides international mobility and cross-sectoral experience through a well-aligned research and training programme. The academic expertise underpinning CfACTs is sourced from BU’s world-class National Centre for Computer Animation (NCCA). 

The research fellow will be an employee of BU and embark on a two-year programme of world leading research with industrial partner Framestore, applying Machine Learning to Solve VFX Light Transport Situations.

High end visual effects production demands ever more convincing and complex imagery. This often includes difficult to solve light transport situations such as nested dielectric media, volumetric multiple scattering and complex lighting scenarios. Typical approaches such as forward path tracing struggle with many important transport types, and computation times for reaching acceptable levels of variance can be significant. Working within the in-house renderer Freak, the researcher will prototype methodologies that use machine learning to provide progressive, controllable and temporally coherent acceleration of light transport across production datasets. In order to be applicable to production rendering, important considerations include temporal coherency, scalability and user control. Key areas of interest include transport through highly scattering media, such as clouds and white fur, and efficient computation of specular-diffuse-specular paths.

This work has the potential to have a direct and significant impact on the quality and efficiency of rendered imagery in the production environment. As well as providing a practical state-of-the-art approach for production quality visual effects rendering, this allows a new level of fidelity and realism to be reachable within visual effects.

Please see the BU‑CfACTs webpage for guidance for applicants:

The CfACTs fellowships follow the specification and requirement of the Standard fellowships of the MSCA; available in full from:

To be eligible to apply an applicant must meet the requirements outlined and:

Be in possession of a doctorate or have at least four years full-time equivalent research experience.
Not have resided or carried out his/her main activity (work, studies, etc.) in the United Kingdom for more than 12 months in the three years immediately before the call deadline.

An applicant will provide a research proposal to outline their research project and an academic CV.

As a part of the application all applicants are required to submit the following:

  • BU Application Form and the Equality Monitoring form
  • Research Proposal (max.10 pages)
  • CV (max.5 pages)
  • Ethics issues - Applicants must provide information on how they intend to address potential ethics issues, as required by the Horizon 2020 Guidance on Ethics Self-Assessment;
  • Signed support letters from the applicants PhD academic supervisors; (optional)
  • Criminal Record Declaration Form

Applications should be sent to

This position is available on a 2 year fixed-term basis.  For additional information please contact the BU Academic Supervisor Dr Richard Southern or Prof Jian Chang, CfACTs Project Coordinator, via quoting FMC232 in the subject line.

Interviews will be conducted for FMC232 in October 2021, exact date to be confirmed. 

BU values and is committed to an inclusive working environment.  We seek a diverse community through attracting, developing and retaining staff from different backgrounds to contribute to inspirational learning, advancing knowledge and enriching society.  To support and enable our staff to achieve a balance between work and their personal lives, we will also consider proposals for flexible working or job share arrangements.