The Digital Emily Project: Achieving a Photorealistic Digital Actor

For more than three decades, computer graphics researchers have been working to create a photorealistic digital actor. In the Digital Emily project, Image Metrics and the University of Southern California's Institute for Creative Technologies (USC ICT) animated a digital face using new results in 3D facial capture, character modeling, animation, and rendering. The project aimed to cross the "uncanny valley" that divides a synthetic-looking face from a real, animated, expressive person. The key technologies included a fast high-resolution digital face-scanning process using USC ICT's Light Stage capture system and Image Metrics' video-based facial-animation system. The project generated one of the first photorealistic digital faces to speak and emote convincingly in a medium close-up.LED lights turned on. We've described the Light Stage scanning process elsewhere. Unlike earlier Light Stage processes, which photographed the face under hundreds of illumination directions, this newer process requires only 15 stereo photographs of the face under different lighting conditions to capture the face's geometry and reflectance. The small number of photos lets us use off-the-shelf high-quality still cameras (Canon ID Mark Ill's) and shoot each data set in under three seconds using the cameras' burst mode.