Government-funded boffins have collaborated with others from famous movie industry labs to produce software which can "clone" a person's facial mannerisms. The cloned data can then be applied to a CGI head, either to let a computer masquerade as a real person or to enhance the realism of animated characters.
“Spoken words are supplemented with non-verbal visual cues to enhance the meaning of what we are saying, signify our emotional state, or provide feedback during a face-to-face conversation,” says Dr Barry-John Theobald, computing science prof at the University of East Anglia.
Theobald has teamed up with other researchers including Dr Iain Matthews, who (it says here) works for both Disney Research and pioneering Kiwi rubberface-movie lab Weta - responsible among other things for Gollum et al in the recent blockbusting Lord of the Rings trilogy. The boffins say their software can extract a "clone" of a person's mannerisms, expressions and facial behaviour from a video of them talking.
According to the UEA:
These facial expressions and head movements can be manipulated live to alter the apparent expressiveness, identity, race, or even gender of a talker. Moreover, these visual cues can be manipulated such that neither participant in the conversation is aware of the manipulation.
As Theobald's team describe it, "we might wish to dissociate behaviour (facial expressions) and appearance (identity)". In other words, once you had an accurate record of someone's facial appearance, you could potentially use "cloned" expression/mannerism records to create an animated simulacrum indistinguishable from the real person. Alternatively, you could imbue a different modelled face/head with the same behaviour as someone else.
“This exciting new technology allows us to manipulate faces in this way for the first time, says Theobald.
"Many of these effects would otherwise be impossible to achieve, even using highly-skilled actors.”
Theobald and his collaborators say that "the new facial expression cloning technique is already being trialled by psychologists in the US... it is also likely to have application in the entertainment industry where life-like animated characters might be required".
The boffins' paper Mapping and Manipulating Facial Expression is published in the June issue of the journal Language and Speech. There's also some illustrative video here.
The research was funded by the Engineering and Physical Sciences Research Council and The National Science Foundation (ie by the UK and US governments). ®