New Scientist reports, “digital motion capture could soon be within reach of low-budget film makers thanks to new software that records movement without using markers.” The system, developed at Stanford University, in California, and the Max-Planck Institute Informatik in Saarbrücken, Germany, allows for what they say is a “3D digital clone” of the actor using a laser scanner. avoiding the use use of body markers or special suits; it also allows the software to capture not only body movement, but everything from facial expressions to clothing. (See demonstration image above and video below.)
This sort of advance seems inevitable and can only add to the anguish felt by many in animation, who as one of my students, Paul Krause, pointed out, seems to reduce animation to a postproduction process. It is also the type of advance that might further encourage live-action filmmakers to move into animation.
The problem with motion capture animation, as with rotoscoping, is not so much the process, but how it is used. Despite all the hype, raw mocap data is rarely useful for animated films, as is and it often requires the assistance of a skilled animator to make it useful; it is this extra step that generally inflates the cost of motion capture, just as it inflates the cost of rotoscoping. (Remember, Max Fleischer gave up the original rotoscoped version of Koko the Clown in 1925 in favor of a more cartoony one because of cost.) While the new process is certainly an interesting breakthrough, it really does not change the basic role that mocap will play for the immediate future, except for perhaps making it somewhat less expensive and perhaps less complicated. The new approach will certainly be welcomed by those who use mocap for video games and scientific purposes, as well as those in visual effects.
In the meantime, go here for a higher resolution, downloadable copy of the video and a preprint of the paper the developers will be presenting at the August SIGGRAPH conference in Los Angeles.