SmartROTO

"SmartROTO is a 21 month R&D project which aims to design, develop and demonstrate intelligent tools for high-end rotoscoping of live action footage and will also investigate the sharing of machine learning networks and datasets between industrial and research communities. The ability to share huge datasets from real world productions has the potential of significantly accelerating the research and development of machine learning.

In modern visual effects ('VFX'), the manual separation of filmed elements, ""rotoscoping"", is an essential part of the digital post-production process. It is time-consuming, manual and largely non-creative, typically outsourced (outside the UK) and has remained unchanged for the last decade. The SmartROTO tools will address the need for optimisation of this process by using machine learning to create a step change in rotoscoping technology. The technology will minimise the interaction required by the rotoscoping artist while keeping a familiar and intuitive user experience.

SmartROTO brings together three world leading organisations: Foundry, Double Negative (DNEG) and University of Bath (UoB). Foundry is the project leader and its success has been built on its cutting edge software development for post-production. In SmartROTO they will provide their expertise in image manipulation tools to develop technology to allow easy rotoscoping of media objects. DNEG is one of the world's leading visual effects, animation and stereo conversion companies for feature film and television. They contribute huge datasets to the project along with their highly experienced technical R&D Team and rotoscoping artists. The University of Bath provide the core innovation in computer vision and machine learning to create novel technologies (algorithms and mathematical models) that will underpin SmartRoto.

The SmartROTO technology will bring significant savings to the VFX industry, a new plugin for NUKE from Foundry and new ways to share training datasets."