Communications of the ACM
Who Wants to Show p as Gandalf at Their Next Meeting?
Researchers at the University of Southern California have developed a novel head-mounted display that enables complex facial and realistic speech animations that until now were the realm of A-list digital artists.
Credit: Allan Davey
University of Southern California (USC) researchers say they are developing novel ways of capturing micro-expressions in the face that are essential to creating natural dialogue and nuanced actor-specific emotions.
The researchers have developed a new head-mounted display that enables complex facial and realistic speech animations. "We want to let people unleash their creativity and come up with their own ways to beat boredom during office meetings," says USC professor Hao Li.
Last year, the researchers say they built the first face-to-face interaction between virtual reality (VR) users wearing head-mounted displays in collaboration, and now they have developed a solution that enables realistic speech animation using deep neural networks.
The new wearable system is based on FOVE, an eye-tracking VR headset, which uses integrated cameras to record eye motions and a mouth camera to track the lower part of the face.
The researchers think this technology will be available to the general public in the next year or two, enabling users to attend meetings and events as characters from movies.
From USC News
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found