Watch These Mesmerizing 3D Printed Motion Sculptures of Lebron James's Dunk
As the years go on, athletes are pretty much always getting better. From their nutrition to their equipment to their ability to study and review their moves, the boundaries of what’s possible rarely, if ever stop moving. Lebron may not have surpassed MJ just yet, but he’s already clocked more games and is still going strong.
Unfortunately for Lebron, but fortunate for his successors, the pursuit of human excellence may have gotten a little easier this week, thanks in part to a new effort from researchers at MIT’s Computer Science and Artificial Intelligence Laboratory to make it much easier to more precisely visualize human motion. These motion sculptures, or “mosculpts,” make it much easier to record and study the finer points of a dunk, a leap, or a tennis serve.
Slow motion video may have accomplished a lot in this regard, but it’s still two dimensional. This erases a lot of the nuance to what makes a Lebron dunk a Lebron dunk, or what makes Roger Federer’s serve so special. You’re also stuck with whatever angle the camera caught in the frame, whereas 3D motion visualizations enable you to rotate and study a given movement from different viewpoints.
Why Visualized Motion Could Be a (Sorry) Game-Changer
Among other things, researchers see an opportunity for the technology to help novices emulate the greats.
“Imagine that you have a video of Roger Federer serving a ball in a tennis match, and a video of yourself learning tennis,” Xiuming Zhang, lead author of a related paper that will be presented next month in Berlin, said in a statement. “You could then build motion sculptures of both scenarios to compare them and more comprehensively study where you need to improve.”
It’s interesting to think about the possible ramifications, not just in sports but in the arts. Dance and the ballet, for example, has always had problem with ephemerality, at least relative to the art forms that can be more readily recorded using text, images, or musical notation. Dance lives in a moment, but motion visualization may change that by giving us a much better tool for preserving, say, Misty Copeland’s pirouette for posterity.
But that may only beginning, too. Right now these visualizations can only be achieved with one key subject, which limits the applications somewhat. Being able to create visualizations with multiple subjects, the researchers say, could open the door to studying all kinds of other situations, like social disorders and team dynamics.
Best of all, this new process is simple. You don’t need a special set or depth cameras in order to create a motion sculpture, you only need an input video. The system then analyzed the video, detects your body parts, and stitches together the best poses from key body parts.