Institute of Neuroscience & Psychology (INP), University of Glasgow
To precisely and objectively characterise dynamic facial expression signals used for social interaction
Development of a unique research facility.
Providing a competitive edge to advance knowledge.
Publication in high profile journals.
Based in the INP at the University of Glasgow, the research team’s goal is to understand the complexities of human social communication - how humans send and receive signals to achieve mutual understanding. One of the most powerful social signals is a facial expression, where specific face movements are made to send a particular message. By understanding what these signals are and the messages they send, the team can understand exactly how social communication is achieved, and perhaps most importantly why communication can breakdown (e.g. between cultures).
The team’s main challenge was to precisely and objectively characterise dynamic facial expression signals used during social interactions, to then identify which signals support accurate communication and those that produce confusions (e.g. between cultures).
The team observed demonstrations of the DI4D PRO System in action and believed it could be used to fulfil their requirements. They used the System to build a database of standardised facial movements, (‘Action Units’), each captured in 4D, (using high resolution digital video cameras) and tracked using the Company’s software.
Using this database of face movements, they created the ‘Generative Face Grammar’ (GFG) - a facial animation system that can produce all facial behaviours by combining different dynamically activated ‘Action Units,’ and provides feature landmark tracking and mesh points in 3D over time - an essential feature for the team.
Professor Philippe G. Schyns, Institute of Neuroscience & Psychology, Director
Dr Oliver G. B. Garrod, Institute of Neuroscience & Psychology
The GFG is a worldwide unique state-of-the-art research facility, and gives the team, and their international collaborators, a competitive edge to advance knowledge. As a result, they have made significant advances in several areas. For example, they revealed for the first time culture-specific differences in facial expressions of emotion - thereby refuting long held notions of universality. They also showed that specific facial movements can mask the default social impression of a face based on its morphology. Their work on dynamic facial expression signals has been published in a number of high impact academic journals including Current Biology, PNAS, and Psychological Science, and featured in various high profile public forums, including New Scientist, Time Magazine, and The Wall Street Journal.
In future, they plan to integrate the data they have accumulated from using their DI3D and DI4D systems with real-time analysis of spontaneous facial expressions from video captures.
Michael Illingworth, Owner, Vine FX
Experience high fidelity facial performance captureRequest a Quote