Active Arts, Culture & Design Computing & AI

CoSTAR Realtime Lab

In plain English

AI plain-English summary

A new lab in Dundee and Edinburgh will combine video game engine technology with film and TV production to create virtual sets and characters in real time, rather than adding them in post-production. The problem is that high-end virtual production—used in films like *The Mandalorian* and for live events—remains expensive, slow, and accessible only to large studios. Smaller UK creative companies lack the tools and pipelines to adopt these technologies. The lab aims to close that gap by developing open, efficient, and sustainable production workflows that any content creator can use. If successful, the CoSTAR Realtime Lab could transform how entertainment is made. Instead of building physical sets or spending months on visual effects, filmmakers and game developers could generate photorealistic environments and virtual actors on the fly. This would cut costs, reduce waste, and allow smaller studios to compete. The lab will also explore machine learning for automating repetitive tasks like lighting and background generation, and develop new standards for ray tracing and 3D scanning. Beyond entertainment, these tools could feed into training simulations, live events, and location-based experiences. The project builds on existing R&D networks in Dundee and Edinburgh, aiming to create a permanent dual-city cluster that generates new companies, products, and global leadership in real-time content creation.

View original technical description
Realtime and virtual graphical environments have been evolving with increasing speed over the last decade. As consumers we are becoming increasingly familiar with high fidelity virtualisation in film/TV and animation, video games and virtual reality. We have seen these techniques and technologies migrate from highly specialised and experimental uses to mainstream adoption and application. The use of these techniques in mainstream film and high-end theatre production has accelerated technological innovation and application across a broad range of motion and performance capture, real-time graphics processing, network and telecoms, edge computing in content and experience creation for entertainment media and live events. At the heart of these innovation is the development of real-time integration technologies primarily used for the creation of video games. Game engine technologies such as Unreal Engine and Unity are designed to integrate dynamic graphical environments with events and environmental effects in real time. For the last 5 years R&D-led growth and innovation in these areas has been driven through the AHRC Creative Industries Clusters InGAME in Dundee and Creative Informatics in Edinburgh. Both programmes have established strong, trusted networks and effective processes for driving demand led R&D into commercialised products. The CoSTAR Realtime Lab to be based in Water's Edge studio complex in Dundee and ECA in Edinburgh will bring together video games development expertise in Dundee's globally significant games cluster with world leading applied R&D at Abertay University and film, television and animation production in Edinburgh with University of Edinburgh's world leading expertise in machine learning and artificial intelligence. The CoSTAR Realtime Lab will deliver close-to-market innovation for digital productivity tools, creative production pipelines and content creation processes. The project will also bring together the existing networks, product development insight and R&D power from InGAME and Creative Informatics clusters to form a dual city creative technologies cluster supported by innovative product and service platforms that will generate new companies, products and markets for high-end, efficient, sustainable and accessible content creation. CoSTAR Realtime Lab will focus on applying innovative technology and process in entertainment contexts initially focussing on driving value through innovation in the following areas. - Performance and motion capture. - Virtual humans and dynamic procedural performance. - Machine Learning for production and Artificial Intelligence for process reproduction. - Artificial intelligence for dynamic effects and procedural graphics for visual effects. - Using advanced scanning technologies for 3D volume acquisition and ML/AI for procedural environment generation. - Developing lighting and ray tracing standards to deliver environmental fidelity. - Location based Software and hardware integration. - Integrated virtual and real-world film and TV production. The vision for the CoSTAR Realtime Lab will be to establish a mixed reality R&D centre that supports sector engagement in collaborative R&D led innovation. The Lab will facilitate engagement between multinational technology vendors and UK content creators with creative technologists and digital content SMEs to identify opportunities for innovation driven growth in scale and value for the Scottish and UK sector. Collaborative R&D will promote product innovative in SMEs by supporting new product, process and service development, driving value and stimulating diversification of product and markets. By combining world leading research with multinational technology companies and highly creative and dynamic SMEs, the CoSTAR Realtime Lab will transform the creative media industries in Scotland and the UK and deliver global leadership in next generation entertainment creation and technology innovators.

View the original record at the funder ↗

Researchers

Andrea Szymkowiak (Co-Investigator)Chris Speed (Co-Investigator)Erin Hughes (Co-Investigator)Gareth Robinson (Co-Investigator)Gregor White (Principal Investigator)Laura Sevilla-Lara (Co-Investigator)Matthew Bett (Co-Investigator)Melissa Terras (Co-Investigator)Phillip Vaughan (Co-Investigator)Xiaoxuan Lu (Co-Investigator)

Related Research

Grants with similar aims, by meaning.

CoSTAR Screen Lab
InGAME: Innovation for Games and Media Enterprise
CoSTAR Live Lab
CoSTAR National Lab
CoSTAR Foresight Lab

Original classification

Research Grant

Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.