May 1, 2007 3-D effects in the movies can now be created by capturing reflected light off actual human faces. Digital images of the strobe-light pictures are animated to produce more realistic human expressions.
A character's movement can look very real in movies, but animators admit it is hard to make the faces look as expressive as human faces. Animators are now working hard on new technologies to make it almost impossible to tell the difference between cartoons and real-life. Ken Pearce, a computer scientist with Mova Contour, admits, "We recognized that facial animation is really one of the last big challenges of computer animation."
Steve Perlman, the inventor and president of Mova Contour, is taking on the animation challenge. His new technology, called Contour, captures facial movements of actors and turns them into very realistic, animated humans. His goal is to make his computer-generated characters indistinguishable from real people. He paints glow-in-the-dark makeup over actor's faces in order to capture striking and subtle facial details from muscles, wrinkles and expressions that human faces make.
After the makeup is applied, the actors go into a studio where lights flicker on and off at speeds too fast for the human eye to notice. In the dark, cameras snap digital images of the glowing face. Using computers, the images are then edited -- inserting fake eyes and teeth where makeup can't be applied. Perlman explains, "When we have humans brought into a computer-generated world, it gives us the flexibility to create any world we want."
The technology is catching on quickly; video game companies want the technology to make more realistic characters, and motion picture companies have orders in to start creating real-life animated worlds. The new reality capture technology is being used in for the upcoming film, The Curious Case of Benjamin Button in which Brad Pitt plays a character who ages in reverse! The movie is scheduled for release in May 2008.
IEEE-USA contributed to the information contained in the TV portion of this report.
BACKGROUND: Contour is a futuristic camera system that will add photorealistic 3D effects to digital entertainment. The system could change the nature of cinematography in several ways. For instance, the system could create compellingly realistic synthetic actors by capturing the facial movements of real actors in much greater detail we currently see. Ultimately, it may help to create a new form of digital video in which the viewer can control the point of view: "navigable entertainment."
WHAT IS MOTION CAPTURE: Motion capture cuts the costs of computer animation while creating more natural movement. Such systems work by tracking the locations of hundreds of reflective balls attached to a human actor. This permits the actor's movements to be sampled by a camera many times per second. But the digital record is limited to movements and does not include the actual appearance of the actor. They are limited in resolution to several hundred points on a human face.
ABOUT CONTOUR: Contour can recreate facial images at a resolution of 200,000 pixels. This vast improvement in resolution will enable filmmakers not just to grab points on a face, but also to capture the look of the entire skin, capturing all the detail and quirks of human expression with startling realism. There are still a few limitations. The Contour system can capture eyebrows, mustaches and short bears, but not freely moving strands of hair. It also cannot capture any areas where makeup can't be applied, like the eyes or the inside of the mouth. Contour's developers are experimenting with plastic teeth molds with embedded phosphor powder to overcome the latter challenge.
HOW CONTOUR WORKS: The Contour system requires actors to cover their faces and clothes with makeup containing phosphorescent powder that is not visible under normal lighting. In a light-sealed room, the actors face two arrays of inexpensive video cameras that are synchronized to simultaneously record their appearance and shape. Scenes are lit by rapidly flashing fluorescent lights, and the cameras capture light from the glowing powder during intervals of darkness that are too short for humans to perceive. The captured images are then transmitted to an array of computers that reassemble the 3D shapes of the glowing areas. These can then be manipulated and edited into larger digital scenes using sophisticated software tools.