C-CATS

Research | Development | Production

Empathic Avatars

Empathic Avatars is part of on an-going initiative that uses the latest Machine Learning, Computer Vision and CGI rendering technologies to develop responsive digital human likenesses. Our aim is to create lifelike virtual actors that can convincingly react to external stimuli through believable facial expression, individuated to both the specific physical characteristics of the generated face and to the psychological and behavioural traits selectable or designable by the user/artist/director.

Key People

Partners

CARV3D


ASC: Automating the Subjective Cineaste

ASC explores the development of a smart archiving and research tool which can analyse and identify cinematic style and genre across a vast data space of cinema and television materials. This tool offers a set of novel searching solutions that transcend current categorisation and tagging techniques, and identifies new ways of analysing and comparing film and television drama material drawn from extensive historical and geographical domains.

Key People

Partners

British Film Institute


REACT

Working in collaboration with the BFI, REACT takes a Deep Learning approach to developing a theoretical and practical understanding of cinema language across its many forms, genres, variances and evolutions over the past century. REACT uses computer vision and machine learning techniques to compare, contrast and ‘learn’ patterns of filmic expression drawn from over a century of moving image creation from around the world; and from there assist in the creation of a diversity of novel cinematic techniques built on the extensive collective viewing experience that only a computational approach can enable.

Key People

Partners

British Film Institute